<%BANNER%>

Extending Dysarthria Research with a Measure of Communicative Effectiveness


PAGE 1

EXTENDING DYSARTHRIA RESEARCH WITH A MEASURE OF COMMUNICATIVE EFFECTIVENESS By NEILA J. DONOVAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

To those who taught me to love learning

PAGE 3

iii ACKNOWLEDGMENTS I wish to thank my mentor Jay Rosenbek (counselor, teacher, master clinician, and friend) for his tireless efforts in all of these ro les. I am grateful to him for seeing what I have to offer and for encouraging me to set my own direction. I appreciate him for being patient when he could have been otherwis e. He leads by example; watching him I understand what it means to be both clinician and scientist of the highest caliber. I thank my committee members (Craig Velozo, Mary Ellen Young, and Diane Kendall) for giving me their time and the benefi t of their expertise. Their insights have been rewarding and their encouragement gra tifying. I believe this committee should be recognized as an example of the ideal committee. They remained committed to my doctoral education and progress from the first meeting to the dissertation defense. They always convened in the spirit of making good science through intelle ctual exchange and mutual respect. They helped shape me into a budding scientist who values rigorous and sometimes challenging scientific discussions among colleagues. I thank the people who helped me complete this dissertation in a timely fashion. They include Nan Musson, who served as my VA liaison and co-princip al investigator at the Malcom Randall VA. She smoothed the way for me at every turn. I thank the individuals who volunteered to transcribe the speech samples, Elizabeth, Summer, Jennifer, and Jessica. I thank Michael O kun, Kelly Foote and Hubert Fernandez, the physicians at the UF Movement Disorders Cent er who allowed me to have access to their patients during both survey development and di ssertation data collect ion. Their entire

PAGE 4

iv team is outstanding, but I want to thank Sara h Munson and Chuck Jacobsen in particular for all their help in gleaning the database Finally, although I cannot mention them by name I thank all of the indivi duals who volunteered to particip ate in this study. They had to open their homes to me and give of thei r time and effort. Everyone was cordial and hospitable. I will treasure the personal stories they shared with me. I could not have made it to this point in my life without the love and support of family and friends. I cannot begin to name ev eryone who has helped me get to this point in my life and to the culmination of this fantastic experience calle d graduate education, for fear of leaving someone out. I was mo st fortunate to get good genes and solid, Midwestern nurturing that gave me a love of reading, a steadfast work ethic, and the desire to help others. I ha ve been doubly blessed to find mentors and friends at every turn on my educational and professional path. I wish to thank them all. Finally I wish to acknowledge my husband Michael Hammer, wh o has always encouraged me to follow my dreams.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 Statement of the Problem..............................................................................................1 Specific Aims................................................................................................................4 Research Hypotheses....................................................................................................5 2 LITERATURE REVIEW.............................................................................................6 World Health Organization Internati onal Classification of Functioning and Disability...................................................................................................................6 Assessment and Treatment of the Dysart hria of Parkinson’s Disease (PD)...............11 Measurement in Dysarthria.........................................................................................16 Operationalizing Communicative Effectiveness........................................................18 Development of the Communicative Effectiveness Survey (CES)............................21 3 DESIGN AND METHODS........................................................................................26 Participants.................................................................................................................27 Recruitment.................................................................................................................29 Parkinson’s Disease Group..................................................................................29 Non-PD Group.....................................................................................................29 Significant Other Group......................................................................................29 Sample Demographic Data.........................................................................................30 Informed Consent.......................................................................................................30 Screening Measures....................................................................................................31 Dependent Measures...................................................................................................33 Procedures...................................................................................................................33 Screening Procedures..........................................................................................33 Dependent Measure Co llection Procedures.........................................................34

PAGE 6

vi Scoring Dependent Measures Procedures...........................................................35 Data Collection Methodology.....................................................................................35 Inter-rater Reliability...........................................................................................36 Data Analysis Plan......................................................................................................37 4 STUDY RESULTS.....................................................................................................39 Results of Testing Hypothesis 1.................................................................................39 Results of Testing Hypothesis 2.................................................................................39 Results of Testing Hypothesis 3.................................................................................41 Post Hoc Analyses......................................................................................................42 Participant Feedback...................................................................................................45 5 DISCUSSION.............................................................................................................47 Hypothesis 1...............................................................................................................47 Hypothesis 2...............................................................................................................49 Hypothesis 3...............................................................................................................51 Improving the Communicative Effectiveness Survey................................................56 Conclusion..................................................................................................................58 APPENDIX A COMMUNICATIVE EFFECTIVENESS SURVEY.................................................60 B HOEHN AND YAHR STAGING FOR PARKINSON’S DISEASE........................62 C PARTICIPANT FEEDBACK ON THE CES............................................................63 LIST OF REFERENCES...................................................................................................66 BIOGRAPHICAL SKETCH.............................................................................................74

PAGE 7

vii LIST OF TABLES Table page 2-1 SIP communication and so cial interaction items.....................................................15 2-2 Demographics of participants in two Rasch analyses of CES ratings......................23 2-3 Results of two Rasch analyses of CES ratings.........................................................24 2-4 Summary of measured steps 4point rating scale PD sample................................25 2-5 Summary of CES 4-point Rating Scale Characteristics of PD Sample....................25 3-1 Inclusion/exclusion criteria for study participants by group....................................28 3-2 Demographic data for all groups..............................................................................31 3-3 Dysarthria Rating of Severity...................................................................................32 3-4 Inter-rater reliability.................................................................................................37 4-1 Summary of variables SIT and SSI contributing to change in CES.........................41 4-2 Descriptive statistics for CES, SIT, SSI, H&Y & DRS...........................................44 4-3 Summary of variables SIT, SSI, H&Y, and DRS contributing to change in CES...44 C-1 Participant feedback.................................................................................................64

PAGE 8

viii LIST OF FIGURES Figure page 4-1 Comparison of mean s for PD and NO groups..........................................................40 4-2 Comparison of means for PD and SO groups..........................................................40 4-3 CES ratings for H&Y stages of PD..........................................................................42 4-4 CES ratings for ratings of dysarthria severity..........................................................43

PAGE 9

ix Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXTENDING DYSARTHRIA RESEARCH WITH A MEASURE OF COMMUNICATIVE EFFECTIVENESS By Neila J. Donovan May 2005 Chair: John C. Rosenbek Major Department: Communicative Disorders Researchers often assume but have not s hown that treatment effects obtained in clinic translate to success in an individua l’s societal participation. Testing this assumption in the field of motor speech diso rders is impossible at present because a psychometrically sound measure of communicat ion in societal pa rticipation does not exist. The best the profession can do is a ssess speech intelligibility, a valid and reliable measure of connected speech used to demonstrate change at the clinical level. The Communicative Effectiveness Survey (CES) may be an instrument to extend dysarthria research into so cietal participation, howe ver, as yet it has unknown psychometric properties. We asked three experimental questions to study the face validity and construct validity of the CES: Does the CES differentiate individuals with dysarthria and Parkinson’s disease (PD) from non-PD gender and mean-age-matched individuals?

PAGE 10

x Do individuals with dysarthria and PD ra te their CES higher than their significant others (SO) rate them? Do traditional measures of sentence inte lligibility (SIT) and spontaneous speech intelligibility (SSI) predict changes in CES? A convenience sample of 75 participants was divided into three groups of 25: individuals with dysar thria and (PD);non-PD group; and SOs of individuals with PD. The dependent variables were the CES (all groups), the SIT and SSI (PD group only). Results showed that the non-PD group rated their CES significantly higher than did the PD group. The PD group rated their CES significantly higher than the SO group rated them. The multiple linear regression model for SIT and SSI as predictors of change in CES was not significant (F[2, 22] = 1.97, p = .163). The model accounted for 15% of the variability (R2 = .152). A post hoc multiple linear regression analysis, adding measures of physical disability (Hoehn and Yahr [H&Y]) and dysarthria severity (DRS) was significant (F[4,20] = 6.88, p = .001), and 59% of the variability was accounted for (R2 = .588). H&Y was the largest ( = .599) and only significant predictor of CES ratings ( t = -2.19, p = .040). The results of this study added evidence to the face validity and the construct validity of the CES. However, to ensure that the CES is a valid and reliable measure of communication in societal partic ipation more work needs to be done on the other types of validity and reliability.

PAGE 11

1 CHAPTER 1 INTRODUCTION Statement of the Problem Researchers often assume but have not show n that the treatment effects obtained in the clinic translate to success in an individual’ s societal participation to fulfill life roles. Testing this assumption in the field of moto r speech disorders is impossible at present because a psychometrically sound measure of communication in societal participation does not exist. The best the profession can do is assess speech intelligibility (a measure of connected speech), which has demonstrated its worth as a measure of change at the clinical level. Or we can measure per ceptual, acoustic and p hysiologic changes in speech. However, if the goal of speech th erapy is to enable a person to communicate more effectively to participate in his life roles (what some call functional communication), then measures of communicati on in societal participation are needed. One such measure that is under devel opment is a measure of communicative effectiveness. Models of rehabilitation such as the World Health Organization (WHO) International Classification of Functioning, Disa bility, and Health (IC F), provide us with a rationale for extending treatment (such as speech therapy) and rehabilitation research into the societal particip ation component. Conceptually we propose that extending research into the societal pa rticipation component of the ICF model requires a different measure called communicative effectiveness. We define communicative effectiveness as a person’s ability to actively and efficiently get his message across successfully in his

PAGE 12

2 home and community settings to fulfill life roles (Hustad, 1999; Lomas et al., 1989; Yorkston, Klasner, & Swanson, 2001). Neurologic diseases such as Parkinson’s disease often cause impairments including a class of speech disorders called the dysarth rias. The typical clinical and research approach to the dysarthrias is to measure and treat the signs of impairment, which can include sound imprecision, disturbed rate reduced loudness and pitch variability. Dysarthria results in problems in verbal communication (speech) “due to paralysis, weakness, abnormal tone, or incoordination of the speech musculature” (Duffy, 2005). A principal measure of dysarthria treatment outcome is speech intelligibility, a measure based on the acoustic presentation of a series of increasingly complex speech productions that a listener can decode, from single wo rds to complex sentences (Yorkston, 1996; Yorkston & Beukelman, 1981a, 1981b; Yorkst on, Beukelman, & Traynor, 1984). The tacit assumption of this approach is that changes in speech intelligibility will have equally positive influences on a person’s communication to fulfill life roles; what the World Health Organization (WHO) Inte rnational Classification of Functioning, Disability, and Health (ICF) identifies as th e component of partic ipation (WHO, 2001). However, studies report no clear one-to-one re lationship between a speech intelligibility score and a person’s communication succe ss beyond the clinical treatment setting (Weismer & Laures, 2002; Yorkston, 1996). As the population ages we expect the incide nce of PD to increase worldwide (Van Den Eeden et al., 2003). It has been estimat ed that approximately 70% of individuals with PD eventually develop speech disorders that affect not only them but their families as well. (Parkinson’s disease Society, 2004) Th erefore it is important that we investigate

PAGE 13

3 aspects of communication disorders that are pe rtinent to this group at the level where the difficulties are the most troublesome, in fulf illing their life roles—th e ICF’s participation domain. The Communicative Effectiveness Survey (C ES) is being studied in our laboratory to serve as a measure of communication in th e ICF societal participation component. The CES originated from an aphasiology ques tionnaire, the Communicative Effectiveness Index, developed by Lomas and colleagues ( 1989). Ball and colleagues (2004) reported on the psychometric properties of an instru ment very similar to the CES (CETI-M) that they used to measure the communicative effectiveness of 25 individuals with amyotrophic lateral sclerosis (ALS ) (one type of movement diso rder that often results in motor impairments and dysarthria). Those au thors demonstrated the usefulness of their communicative effectiveness measure (CETI-M ) to identify social communication needs of persons with ALS and their significant others, people for whom traditional speech therapy is often contra-indicated (Ball et al ., 2004). They also reported that the CETI-M showed psychometric properties including f ace validity and content validity for what today is classified as the pa rticipation component of the ICF. However, because the speech impairments of individuals with ALS make up only one subset of the dysarthrias, research is needed that incl udes individuals with other type s of dysarthria as well. We chose to investigate one of those groups, individuals with hypokinetic dysarthria and Parkinson’s disease. In a prior study we investigated the m easurement properties of the Communicative Effectiveness Survey (CES) using a singleparameter model of item response theory called Rasch analysis. We conducted two anal yses: first for 95 indi viduals with various

PAGE 14

4 movement disorders and speech competency from normal to severe dysarthria; and second for 66 individuals with idiopathic PD and speech competency from normal to severe dysarthria. Rasch analysis gives information that can be compared to the traditional test-theory terms such as validity, reliability and responsiveness. Our results indicated that the CES demonstrated suffi cient measurement properties including: the theoretical hierarchy matched the sample hierarchy of items (similar to face validity), individuals with movement disorders and arrange of speech competence endorsed the items on the survey in a predictable manner (similar to content validity, and there was high (.90) person-to measure correlation (com parable to Cronbach’s alpha), similar to traditional measures of reliability of res ponse (for both groups). The results were considered strong enough to prompt us to cont inue investigating its usefulness as a tool that we could use to measure Communication at the level of societal participation. We used the data from the earlier analyses to design this study to add further validity to the instrument for individuals with PD. The goal of this study was designed to a dd evidence to the face validity, content validity and construct validity of the Communicative Effectiven ess Survey for individuals with dysarthria and PD by accomplishing three specific aims. Specific Aims Specific aim 1: Determine if CES ra tings differentiate normal speakers from dysarthric speakers with PD. Specific aim 2: Determine the difference in CES ratings by participants with dysarthria and PD and th eir CE ratings completed by significant others. Specific aim 3: Determine the relations hip between two measures of speech intelligibility and CES ratings of participants with dysarthria and PD.

PAGE 15

5 Research Hypotheses Based on the established specific aims we developed three experimental hypotheses. Hypothesis 1: The CES ratings for 25 part icipants with dysarthria and PD will be significantly lower than CES ratings of 25 mean genderand age-matched healthy participants. Hypothesis 2: The CES ratings of 25 part icipants with dysarthria and PD will be significantly higher than the CES ratings given to them by their significant others (SO). Hypothesis 3: When predicting senten ce intelligibility score (SIT) from communicative effectiveness survey score (C ES), spontaneous speech intelligibility score (SSI) accounts for a si gnificant amount of the va riation in SIT above and beyond the variation due to CES for individu als with dysarthria and PD. Likewise, when predicting SSI from CES, SIT accounts for a significant amount of the variation in SSI above and beyond the variation due to CES for individuals with dysarthria and PD.

PAGE 16

6 CHAPTER 2 LITERATURE REVIEW To understand the development of our thesis that dysarthria rese arch needs to be extended into the ICF societ al participation component, we divided review of the literature into four sections which will incl ude: a description of th e components of World Health Organization’s ICF and how they pertain to research a nd treatment of dysarthria; a brief summary of the state of the art in assessment and treatment of hypokinetic dysarthria (the dysarthria that results from Parkinson’s disease) ; a review of the traditional treatment outcome measures of dys arthria; and an operational definition of communicative effectiveness; and a review of how the CES has been tested in other laboratories. The final section of the ch apter gives the foundational evidence that the CES had sufficient measurement properties to continue investigating its validity as a measure of communication in societal particip ation for individuals with idiopathic PD and dysarthria. World Health Organization Internationa l Classification of Functioning and Disability The WHO International Classi fication of Functioning, Disa bility and Health (ICF) is one of the models used in the rehabili tation literature (WHO, 2001). The ICF is a compelling conceptual model for two reasons First, the ICF’s biopsychosocial design attempts to integrate the most important aspe cts of both the medical model and the social model of functioning and disabi lity. The ICF defines health -state as a bi-directional dynamic process involving the person with the disease/illness, which can be positively or negatively affected by contextu al/ environmental and person al factors. It is now

PAGE 17

7 recognized that a constellation of environm ental and personal factors may positively or negatively impact health-state and functioning, regardless of the type or intensity of treatment received (Cardol et al., 2002; Jette, 2003; Keysor & Jette, 2001). Second, the ICF is a compelling model because it illustrate s the complexities involved in planning and carrying out investigati ons and interventions to achieve and maintain optimum functioning of individuals in compromised health states (Brandt & Pope, 1997). The ICF includes 2 components: body stru ctures and functions; and activity and participation (previously know as the domains of impairme nt, handicap, and disability) (WHO, 2001). Activity is define d as a discrete task or acti on that a person executes. Participation is defined as a complex task that may require multiple sets of skills to complete. Chapter 3 of the ICF describes the activity and participati on skills needed for communication. A communication activity for speech production is “producing words, phrases and longer passages in spoken message s with literal and implied meaning, such as expressing a fact or telling a story in oral language” (p. 134) (WHO, 2001). Communication for societal part icipation is defined as “hol ding a conversation: starting and sustaining an interchange of thoughts and ideas, carried out by means of spoken, written, sign or other forms of language, with one or more people one knows or who are strangers in formal or causal settings” (p. 135) (WHO, 2001). Impairment is the term used to denote a deficit in body structures an d/ or body functions (i.e paralysis of a body as a result of cerebral infarct). Activity limita tions and participation restrictions are the terms used when an individual cannot successfu lly complete the delineated skills under the conditions defined by the ICF.

PAGE 18

8 The inclusion of these domains, allows for the possibility of research to differentiate between an individual’s capacity to perform an activity and his actual participation in fulfilling societal roles. However, there ar e now published reports by investigators that many different types of raters including healthcare providers and rehabilitation scientists have difficulty differentiating between activity limitation and participation restriction (Jette, Haley, & Kooyoomjian, 2003; Norden felt, 2003; Perenboom & Chorus, 2003). These reports provide initial evidence that th e theoretical distinction between activity and participation may be more unders tandable than the concrete ap plication of rating them in rehabilitation research, assessment, and treatment. The science of rehabilitation is moving towa rd an integrated a pproach that includes consideration of performance, context of performance, and ps ychosocial issues thanks to conceptual models of enablement such as the ICF. Unfortunately, most studies of dysarthria have not kept up with the current advances in rehabilitation (Enderby, 2000). Rather, most studies of dysa rthria are based on perceptu al judgments, acoustic, and physiologic measures. Speech-language pathol ogists make perceptual judgments during an individual’s speech evaluation. Perceptu al judgments are typi cally made using a rating scale to judge the voice quality, pitch, loudness, prosody, intonation, hypernasality, and precision of articulati on of speech. Acoustic measures include fundamental frequency, vocal intensity, maximum duration of phonation, semitone standard deviation. Physiologic measures include things such as forced vital capacity, maximum inhalatory pressure, and maximum exhalatory pressure (Ansel & Kent, 1992; Hammen & Yorkston, 1996; Kent, 2000; Kent & Read, 1992; Kent, Vorperian, Kent & Duffy, 2003; Klasner, Yorkston, & Strand, 1999; Rosenbek & LaPoint, 1978). These kinds of measures are

PAGE 19

9 related most directly to the body structur es/functions component of the ICF (Enderby, 2000). Specifically, the ICF Body Structur es and Functions Chapter 3 “Voice and Speech Structures/Voice and Speech Func tions” describes the body structures and functions required to pr oduce voice and speech. On the other hand, the commonly used assessment of connected speech production, the Assessment of Intelligibility in Dysarthric Speech (AIDS), would be classified in the activity component (Ball et al., 2004; Yorkston & Beukelman, 1981a, 1981b; Yorkston, Beukelman, & Tice, 1996; Yorkston et al., 1984; Yorkston, Dowden, & Beukelman, 1992; Yorkston, Strand, & Kennedy, 1996). For example the ICF Activity/Participation Chapter 3 states that comm unication addresses receiving, and producing different forms of communication including words, gestur es and signs, pictur es, writing, and body language, along with using augmen tative communication devices. Although the ICF provides the framework fo r a more integrated rehabilitation approach in dysarthria research, such resear ch has been hindered by an absence of valid and reliable tools for measuring communi cation beyond the body stru cture/function and activity components of the ICF. Pertaining to the hypokinetic dysarth ria associated with PD, the bulk of the literature would be cl assified in the body structures/functions component of the ICF (WHO, 2001). There is scant documentation of what impact, if any, hypokinetic dysarthria has on an individual’s participation in fulfilling hi s societal roles. However, according to Yorkston and colleagues (2001), “there remains an urgent need to develop measures that reflect the quality of communication particip ation in adults with acquired neurologic communication disorders” (p.136).

PAGE 20

10 We do have an instrument that was deve loped to show change in communicative effectiveness of individuals with aphasia befo re and after treatment. That instrument is the Communicative Effectiveness Index (CETI). We suggest that the CETI has d content validity as a measure of participation. The Communicative Effectiveness Index (CETI) items were developed from interviews that involved individuals w ith aphasia and their caregivers about the everyda y speaking situations wher e individuals experienced decreased communicative effectiveness (Lomas et al., 1989). Additionally, the CETI was designed to show changes in communicative interactions in social situations for individuals with aphasia. The authors reported that individuals with aphasia demonstrated improved communicative effectiv eness after therapy based on changes in ratings of communicativeness (Lomas et al., 1989). Of late the psychometric properties of the CETI-M (an instrument very simila r to the CES) were reported based on the ratings of individuals with dysarthria and amyot rophic lateral sclerosis (Ball et al., 2004). These authors also presented evidence of the face validity of the CETI-M as a measure of communication in societal participation. In conclusion, we suggest that the WHO IC F and other conceptually driven models of functioning and disability provide the rationa le for an instrument that will measure an individual’s communication to succ essfully participate in life roles. We do not have such an instrument at this time in the field of speech pathology. We suggest that the CES may be an instrument that was originally designed to be such a measure and that there is early evidence that it has usefulness for individuals with amyotrophic lateral sclerosis. We do not, however, know how individuals with PD would rate their communicative effectiveness on the CES.

PAGE 21

11 Assessment and Treatment of the Dysart hria of Parkinson’s Disease (PD) Any neurologic disease afflicting the bul bar system can result in a motor speech problem called dysarthria. Parkinson’s diseas e (PD) is such a disease and the resulting motor speech problem is called hypokinetic dysar thria. Parkinson’s disease can affect 500,000 people at any one time (Van Den Eeden et al., 2003). An estimated 70% of individuals diagnosed with PD experience dys arthria ranging from mild to severe over the course of the disease (Yorkston, Miller & Strand, 2004). Dysart hria can affect the person as well as the speech. Consider this statement presented on the Parkinson’s disease Society website: speech disorders: can interfere with all aspects of communica tion and as a result there can be a great deal of unhappiness, frustra tion, and misunderstanding, not only for the person with PD but also for the significant other and the people the person comes into contact with (Parkinson’s disease Society, 2005 http://www.parkinsons.org.uk ). Dysarthria is not a single speech disorder, but several different speech disorders. Duffy (2005) defined the dysarthrias broadly as: the collective name for a group of speech di sorders resulting from disturbances in muscular control over the speech mechanis m due to damage of the central or peripheral nervous system. It designates problems in oral communication due to paralysis, weakness, abnormal tone, or in coordination of the sp eech musculature (p. 5). Specifically, the hypokinetic dysarthria asso ciated with PD has been identified based on the following perceptual characte ristics: reduced intensity (monoloudness), reduced stress and intonation patterns (monotone, monopitch), abnormal voice quality, articulatory imprecision of consonants produ ced in rushed bursts of speech, and inappropriate or illogical pauses duri ng speech (Adams, 1997; Darley, Aronson, & Brown, 1975; Duffy, 1995, 2005; Rosenbek & LaPoin t, 1978; Yorkston et al., 2004).

PAGE 22

12 Ramig and colleagues built a substantial body of literature showing the immediate and long term treatment efficacy of the L ee Silverman Voice Treatment (LSVT) for individuals with hypokinetic dysarthria and PD (Ramig, 1994, 1995; Ramig, Countryman, O'Brien, Hoehn, & Thompson, 1996; Ramig, Countryman, Thompson, & Horii, 1993; Ramig, Countryman, Thompson, & Horii, 1995; Ramig et al., 2001). Lee Silverman Voice Treatment is an intensive therapy (4 hours per week for 16 weeks and daily home practice) based on the theory that training maximum phonatory effort leads to increased vocal intensity that, in turn, ge neralizes to other sp eaking situations. Ramig has reported that changes in one or more acoustic or physiologic measure (fundamental frequency, vocal intensity, maximum duration of phonation, semitone standard deviation, and forced vital capaci ty) and perceptual ra tings (loudness, pitch, monotonicity) were significantly higher for individuals who received LSVT compared to those who received a treatment of respirator y effort at post-treatment, 12 months posttreatment, and 24 months post-trea tment intervals (Ramig, 1994, 1995). If we look at how the authors categorized their assessment tools with regard to the ICF components of body structur es/functions and activ ity/participation, we see that selfratings and family-ratings of the perceptual categories of loudne ss, monotonicity, and hoarseness were described as measures of f unctional communication. Ratings of speech intelligibility and conversational initiation were described as measures of activity limitations (Ramig, 1995). Finally, they cate gorized ratings on the Communication and Social Interaction subtests of the Sickness Im pact Profile (SIP) as measures of functional outcome (Ramig et al., 1996; Ramig et al., 199 5). The SIP (Bergner, Bobbitt, Carter, & Gilson, 1981; Gilson et al., 1975) is defined by its authors as a behaviorally-based

PAGE 23

13 measure of sickness-related dysfunction, ra ther than a functional outcomes measure (Pollard & Johnston, 2001). Ramig group’s use of preand post-treat ment self-report ratings as functional communication outcome measures bears review. Results of these self-report ratings for males and females in both treatment groups resu lted in changes that were statistically significant for all perceptual measures (e xcept loudness) including: monotonicity, hoarseness, intelligibil ity, and initiation of conversation. Self-ratings of loudness resulted in a time, treatment, gender interaction wh ereby males who received LSVT showed the greatest improvement. These re sults showed that from the patients’ perspectives, both treatments resulted in improvement on all va riables rated (except l oudness) for a specific group of males and a specific group of female s. On face value, these results added credence to the theorized variability of se lf-report ratings from individuals PD. For example, other authors reported that individuals with mild to moderate dysarthria and PD had reduced insight into their speech de ficits (Antonius, Beukelman, & Reid, 1995; Yorkston, Bombardier & Hammen, 1994). Other considerations, however, may ha ve to do with study methodology. The authors state that two speech -language pathologists admi nistered treatment to all individuals and “worked closely together to ensure consistenc y and equivalent high effort and motivation across both forms of treatme nt” (p. 1234) (Ramig et al., 1995). It was reported that individuals were asked to complete self-rating forms at different intervals during and after the study. However, the auth ors do not report how the self-ratings were obtained (i.e., did the clinicians administer the self-report fo rm, or were the clinicians present when the individual completed the form). The bias of participants trying to give

PAGE 24

14 responses that pleased the c linicians may have been a c onfound to the results of this study. Another possibility is that the variables were difficult for untrained individuals or their families to rate in the manner intended by the investigators. Unfortunately, in the reported comparisons of short-term (12 months ) and long-term (24 months) results of the LSVT, Ramig et al., (1996) did not report self -ratings or family ra tings of perceptual variables. If these self-ra tings and family-ratings were used as outcome measures (comparing pre-treatment and post-treatment ra tings) it would have been useful to our understanding about the effectiveness of the tr eatment from the subject’s point of view (the functional perspective) to have the resu lts of the same measures at 12 months and 24 months post-treatment. The choice of a measure of sickness-rela ted dysfunction, such as the SIP (Bergner et al., 1981), may also have aff ected their results. The SIP is a widely used measure of dysfunction related to sickne ss (Pollard & Johnston, 2001). Ramig and colleagues (1996) used two subtests of the SIP (Communication and Social Interaction) to measure functional improvement. Individuals who recei ved LSVT rated a significant reduction of the impact of their illness (PD) on the comm unication subtest of the SIP (i.e. the affect that PD had on their communication was re duced), while the placebo treatment group rated no significant change in the impact of sickness on their communication immediately after treatment (i.e., the affect that PD had on their communication was unchanged). At 12 months post-treatment the significant change reported on the SIP by the LSVT group was not maintained, and the placebo group re ported a significant increase in sickness impact on communication (i.e., the affect that PD had on their communication had

PAGE 25

15 increased, communication was more difficult) (Ramig et al., 1996). No significant differences were found on ratings of the impact of sickness on social in teractions (i.e. PD had no affect on their social interacti ons), for either group, at either time. The results seemed contradictory becaus e many of the social-interaction items required communication to participate in the social interactions, and one would expect that if communication had improved there w ould also be improvement in the social interaction items that pertained to comm unication. See Table 2-1 for examples of communication and social intera ction items from the SIP. Table 2-1. SIP communication an d social interaction items Communication Items I do not speak clearly when I am under stress I have difficulty speaking, for example get stuck, stutter, stammer, slur my words I am having trouble writing or typing I react slowly to things that are said or done Social Interaction Items I am cutting down on some of my usual recreation and pastimes I am doing fewer community activities I am cutting down on the length of visits with friends I am doing fewer activities with groups of people (Bergner et al., 1981) Review of the examples of the items on the SIP showed that the communication items address behaviors we would classify in the ICF activity domain, while social interaction items address behaviors we would classify in the ICF participation domain. So although Ramig and colleagues (1995, 1996) us ed the SIP as an outcomes measure for societal participation, we sugge st that their results showed that the LSVT was efficacious at the level of activity lim itation, but not at the level of societal participation. Studies by Ramig’s group comprise the bulk of the treatment efficacy research for hypokinetic dysarthria. Along with the treatmen t efficacy data, the authors reported that the treatment (increasing vocal intensity) led to improved communication skills, which

PAGE 26

16 led to improved quality of life for individua ls with PD, though they provided no objective evidence. We need evidence to substantiate or refute those subjective observations. Ramig and colleagues did not report on the sa me measures in the study of treatment efficacy and treatment effectiveness. They provided only subjective observations regarding improved quality of lif e. With regard to treatment effectiveness they reported that the treatment was effective based on ac oustic and perceptual measures although the SIP communication scale ratings were not main tained at either 12 months post-treatment or 24 months post-treatment (Ramig et al., 1996) If the goal of speech therapy, for exam ple LSVT, is to enable a person to communicate more effectively to participate in his life roles, measures of participation are needed. Presently treatment effectiveness is based on measures that are applicable to the clinical setting, but not to the so cial participation of individuals. Evidence-based practice guidelines would s uggest we should only administer those treatments that improve communication effectiveness in daily life or at the participation level. Unfortunately present measures focu s on clinical efficacy versus participation effectiveness. The results of the LSVT literatu re point to the need for an instrument that can measure the communication of individuals at the level of societal participation. We propose that the CES is such an instrument. Measurement in Dysarthria Kent (2000) reviewed the motor speech li terature from its infancy to 2000, and stated that while there have been instrument al and technological a dvances, research has not moved far beyond the traditi onal studies of the classifica tion and/or diagnosis of the dysarthrias. Darley, Aaronson and Br own (1975) presented a systematic description/classification of th e dysarthrias by perceptual characteristics that continues to

PAGE 27

17 be widely used in 2004. Their work was followed by a point-place model in which the anatomic places where important speech activ ities occur are numbered and assessed by perceptual judgments (Rosenbek & LaPoin t, 1978). Beukelman and Yorkston (1977; 1979; 1981b) posited that while the work of both Darley and colleagues (1975), and Rosenbek and LaPoint (1978) were beneficial for classification and differential diagnosis of the dysarthrias, they did not give qu antification of the person’s overall speech production. Thus they developed and sta ndardized a method for assessing speech intelligibility intended to give a measure of an individual’s speech production of words and connected speech (Yorkston & Beuke lman, 1981a; Yorkston, Beukelman et al., 1996; Yorkston et al., 1984). Speech intelligibility at its simplest is de fined as “the extent to which a listener understands the speech produced by individuals with motor speech disorders.” (p. 486) (Yorkston, Beukelman, Strand, & Bell, 1999). To obtain a speech intelligibility score, a nave judge transcribes a tape-r ecorded sample of words and/ or sentences without benefit of context or visual cues. The result is a percentage of intelligible words from total words possible (Yorkston & Beukelman, 1981b; Yorkston et al., 1984). Changes in speech intelligibility scores are traditiona lly reported to document treatment outcome (Yorkston, 1996). Speech intelligibility is a recognized measure of speech production; however, procedures must be clearly controlled to mi nimize threats to the internal validity of a study. Such threats include listener familiari zation (Hustad & Cahill, 2003; Spitzer, Liss, Caviness, & Adler, 2000), length of utte rance (Spitzer et al., 2000), speaking task (Kempler & Van Lancker, 2002), medication fluctuations (Goberman & Blomgren, 2003;

PAGE 28

18 Goberman, Coelho, & Robb, 2002; Larson, Ra mig, & Scherer, 1994; Mawdsley & Gamsu, 1971; Quaglieri & Celesia, 1977), pr actice (Inzana, Driskell, Salas, & Johnson, 1996), and attention (Ho, Ians ek, & Bradshaw, 2002). We know that sampling speech intelligibili ty under well-controlled circumstances will lead to a reliable measure of an individual’s speech inte lligibility, and can serve as a measure of treatment outcome per the wo rk of Yorkston and Beukelman in their development of the Assessment of Intelligibility of Dysart hric Speech (AIDS) (Yorkston & Beukelman, 1981a, 1981b; Yorkston et al., 1984) However, authors report that there is not a one-to-one relationship between a speech intelligibility score and a person’s communication success outside of the clinic al treatment setting (Weismer & Laures, 2002; Yorkston, 1996). Up to now, the lack of valid and reliabl e tools of functional communication (such as communicative effectiveness) has hindere d the investigation of the relationship between speech intelligibility and functional communication. In a natural setting communicative interactions occu r in a conversational contex t. Communication partners have the benefit of contextual cues, gestur al cues, written cues, picture cues, and if needed, active conversational repair, to assist with comm unicative effectiveness. We concur with Ball et al., (2004) that we need valid and re liable measures to assess successful communicative effectiveness in the participation component. We propose that the Communicative Effectiveness Survey (CES) is such a measure. Operationalizing Communica tive Effectiveness We define communicative effectiveness as a person’s ability to actively and efficiently get his message across successfully in his home and community settings to fulfill life roles, using whatever means of communication possible (Hustad, 1999; Lomas

PAGE 29

19 et al., 1989). Furthermore, we conceptuali ze communicative effectiv eness as a construct to be investigated in the participation domain of the ICF as have othe rs (Ball et al., 2004; Hustad, 1999; Lomas et al., 1989; Yorkston et al., 1999). Several authors have advocated the use of a rehabilitation model such as the WHO ICF in dysarthria research (Y orkston, 1996; Yorkston et al., 2001; Yorkston, Strand et al., 1996). However, only two recent studies reported on the disabling nature of dysarthria in a communicative context from the perspective of people with dysarthria. In the first, a qualitative study of 7 participants with mu ltiple sclerosis and dysarthria, subjects’ participation in social interactions an d communication opportunities was severely affected although they had mild communica tion impairments based on their speech intelligibility scores (Yorkston et al., 2001). This is evidence that individuals focus their concerns at the level of part icipation restrictions, rather than impairment or activity limitations. The authors underscored the need for treatment goals focused on decreasing participation restrictions as the starting point of treatment rather than the ending point (Yorkston et al., 2001). In a second study, Ball and co lleagues (2004) reported th at 25 individuals with ALS and their caregivers rate d the communicative effectiveness of the person with ALS similarly (correlation between ratings from ALS speakers and their caregivers was r2 = .87, p = .00). In addition, both ALS speakers and their caregivers ranked the ten items on the CETI-M similarly with re gard to level of difficulty (Wilcoxon signed ranks analysis, no significant differences). In addition, subjects also rate d their communicative effectiveness in social situations more severely impaired although their speech intelligibility scores were as high as 90%.

PAGE 30

20 These articles indicated that individuals with dysarthria associated with multiple sclerosis and ALS reported that they were restri cted from participation in life roles due to their communication deficits (Ball et al ., 2004; Yorkston et al., 2001). We do not have information of this sort for individuals with hypokinetic dysarthria and PD. In summary, models of rehabilitation provide a rationale to begin developing treatments and outcomes measures at the level of societal participation. The state of the art in dysarthria treatment and outcomes is to use traditional, clinical measures of perceptual judgments, acoustic, and physiolo gic measures to show treatment efficacy. Those measures are often accompanied by a st andard measure of connected speech called speech intelligibility. A speech intelligibility score has worth as a measure of change at the clinical level (termed ac tivity component by the ICF), ho wever it has not been shown that the effect obtained in the clinic translates to success in an individuals’ participation in fulfilling his life roles (ter med participation component by the ICF). It is imperative that research be done at the leve l of participation to identify what variables are critical to successful communication in that domain. We propose that an instrument, such as the CES, which has demonstrated face validity in use with individuals with aphasi a and individuals with dysarthria and ALS, may be such an instrument. However, the CES has been an instrument with unknown psychometric properties relativ e to individuals with othe r kinds of dysarthria and neurogenic disorders, such as those with dysarthria and Parkinson’s disease. We did a series of Rasch analyses to study the measurement properties of the CES before using it as a measure in the curre nt study. The next se ction describes the

PAGE 31

21 development of the CES and the results pert inent to its usefulness as a measure of societal participation fo r individuals with PD. Development of the Communicative Effectiveness Survey (CES) Lomas and colleagues (1989) published a repo rt on the development, validity and reliability of the Communication Effectiveness Index (CETI). To establish face validity for the instrument they enlisted individuals with aphasia and their spouses to generate items for the instrument. Next they tested the CETI using two groups of individuals; one group (n =11) still recovering with aphasia; and one group (n =11) stable with chronic aphasia (>12 months post-CVA). The aut hors reported the following internal validity scores: internal validity wa s high based on Cronbach’s alpha = .90; test-retest reliability interclass correlation = .94 (95% C I = .87 to .91) and moderate inte r-rater reliability interclass correlation =.73 (95% CI = .62 to .81). In addition, construct validity was determined using Spearman rank correlation coefficients between the CETI and three measures of aphasia. Their results indi cated that the CETI showed a change in communicative effectiveness preand post-trea tment for individuals with aphasia. We performed a post hoc analysis of power and effect size ( t = 3.68, df = 20, d = 1.57; 95% CI =.6 1.57 2.5) to determine the statistical rigor of the instrument. Power of .94 for effect size of .8 indicated adequate statisti cal power of the results although the sample size was small. Yorkston et al., (1999) modifi ed Lomas and colleagues’ (1989) CETI from 14 to 8 items to be used by individuals with dysarthri a. Their first modification of the CETI was to delete the items that pertained to language that did not pertain to individuals with motor speech disorders. Next, they involve d individuals with dys arthria in the item selection process. The apparent usefulness a nd face validity of the instrument without the

PAGE 32

22 necessary statistical analyses was the impe tus for two stages of Rasch analysis to establish the measurement characteristics of the CES. Rasch analysis, a single-parameter model of item response theory, was used to determine the measurement pr operties of the CES using computer software called WINSTEPS Ministep (Linacre, 1991). Rasc h analysis is designed to provide an investigator with information about the importa nt constructs of measurement, such as the initial content validity of the instrument, re liability of responses of the participants, unidimensionality of items, along with how we ll the items fit together to define a theoretical construct (such as communicative effectiveness). In addition, Rasch analysis transforms ordinal data (counts) into interval data, which results in a log-linear measure of item difficulty, and person ability ca librated on a single scale (Bond & Fox, 2001; Rasch, 1980; Wright, 1997; Wright & Linacre, 1989). Finally, Rasch analysis generates information on ceiling and floor effects, indicato rs of whether or not the instrument had a sufficient level of difficulty to accommodate the range of the population the instrument was designed to measure (Velozo & Peterson, 2001). In the first stage of analysis to determ ine the measurement characteristics of the CES a speech-language pathologist blind to the purpose and hypotheses of the study administered the CES, an eight-item, seven-poi nt interval scale, to 95 individuals with movement disorders (diagnosed by a movement disorders specialist from the University of Florida Movement Disorders Center). Th e individuals with movement disorders had speech competencies ranging from normal to severe dysarthria. Table 2-2 gives the demographic data for the group.

PAGE 33

23 Table 2-2. Demographics of participants in two Rasch analyses of CES ratings All Movement Disorders) PD only Mean age 62.6 (range 18-89) 66.5 (range 49-87) Gender distribution 58% Male, 42% Female 64% Male, 36% Female Diagnosis PD 70%, Dystonia 10%, Other 20% PD 100% Speech competence Normal to severe dy sarthria Normal to severe dysarthria We hypothesized that the CES would show measurement properties sufficient for use as an instrument to measure the communicative effectiveness of people with movement disorders, as it had with individuals with aphasia. Rasch analysis was used to determine the measurement prope rties of the instrument by c onverting the ordinal data to interval data using WINSTEPS, a computer software program (Linacre, 1991). See Table 2-3 for a summary of the Rasch analysis. The results of the first analysis showed that the CES had sufficient measurement properties, along with initial construct vali dity and response reliability, to continue investigating its usefulness as a measure of communicative effectiveness for individuals with movement disorders and a wide range of speech competence. The only exception to our findings was that individua ls used a 5-point rating scal e more effectively than a 7point scale. The 5-poi nt scale was used for all further analyses. The purpose of the second stage of anal ysis was to analyze the measurement properties of the CES for individuals with PD (70% of the original sample). See Table 22 for the group demographics. We hypothe sized that the CES would show similar measurement properties for individuals with PD, as it showed for individuals with various movement disorders. The results s howed that the second analysis were indeed similar to the first, with the exception that individuals with PD used a 4-point rating scale more effectively than the 7-point scale. See Table 2-3 for a summary of the second Rasch analysis.

PAGE 34

24 Table 2-3. Results of two Ra sch analyses of CES ratings All movement disorders PD only Sample size >50a N=95 N=66 Person-to-Measure Correlation* .90 .90 Unidimensionality (MnSq=1 Ideal) In-fit Statistics Persons In-fit Statistics – Items MnSq = .99 MnSq = .98 MnSq = .99 MnSq = .98 Levels of Difficultyb, c 3.13 statistically significant levels of ability 3.05 statistically significant levels of ability Actual Item hierarchy agreement with theoretical item hierarchy 7/8 (1 item) Hierarchy of items agreed with the hierarchy proposed a priori 6/8 (1 item) Hierarchy of items agreed with the hierarchy proposed a priori Redundancy of items 1 Talks when upset rated the same difficulty as talks in a noisy environment 1 Talks when upset rated the same difficulty as talks in a noisy environment Reliable use of rating units Used a 5 point rating scale more reliably than a 7 point scale Used a 4 point rating scale more reliably than a 7-point scale a(Linacre, 1994), b(Wright & Stone, 1979), c(Velozo & Peterson, 2001)*Comparable to Cronbach’s alpha To determine that the 4-point rating scale was more effective than the 5-point rating scale, we used a set of 8 guidelines pres ented by Linacre (2002). We applied the guidelines to the summary of measured steps to determine the effectiveness of the 4-point rating scale. See Table 2-4. Our results met 7 of 8 of the criteria th at Linacre established as essential for rating scale use. See Table 25 for those results. The exception was that there was not a regular distri bution in our category count. Based on both populations analyzed, the CES showed sufficient measurement properties, along with initial face validity, content validity, and reliability of response (based on person-to-measure correlation, sim ilar to Cronbach’s alpha), to continue

PAGE 35

25 investigating its usefulness as a tool to quantify comm unicative effectiveness of individuals with PD and a wide range of speech competence. Table 2-4. Summary of measured step s 4-point rating scale PD sample Category Count *1, 2 Average Measure *3 Expected Measure *3 OUTFIT MnSq *4 (<2.0) Step Calibration *5 (>1 logit) Coherence M C% *6 (>40%) Coherence C M% *6 (>40%) Zone From *7, 8 (>1.4) Zone To *7, 8 (<5.0) 1 28 -4.02 -4.09 1.03 72% 46% -Inf. -4.50 2 179 -1.07 -1.04 1.03 -4.48 73% 75% -4.50 .49 3 166 1.95 1.94 .89 .57 64% 72% .49 4.00 4 87 4.47 4.45 1.04 3.91 78% 64% 4.00 +Inf. Note: *Number identifies pertinent guideline in Table 2-5 Table 2-5. Summary of CES 4-poi nt Rating Scale Characteristics of PD Sample Guideline Measure Stability Measure Accuracy Description of this sample Inference for next sample Pre. Scale oriented with latent variable (E) MET (E) MET (E) MET (E) MET 1 At least 10 observations of each category (E) MET (H) MET (H) MET 2 Regular Observation distribution (H) UNMET (H) UNMET 3 Average measures advance monotonically with category (H) MET (E) MET (E) MET (E) MET 4 OUTFIT mean-squares less than 2.0 (H) MET (E) MET (H) MET (H) MET 5 Step calibrations (H) MET 6 Ratings imply measures, and measures imply ratings (H) MET (H) MET 7 Step difficulties advance by at least 1.4 logits (H) MET 8 Step difficulties advance by less than 5.0 logits (H) MET (Linacre, 2002) Note: (E) = Essential; (H) = Helpful, Blank areas = not applicable to the guideline

PAGE 36

26 CHAPTER 3 DESIGN AND METHODS This study was designed to test the face validity, content validity, and construct validity of the CES as a measure of societal participation for indivi duals with dysarthria and PD. To accomplish that goal we designed a prospective study with mixed statistical design. To test Hypothesis 1 a case-control, one-directional paired samples t-test, = .05 was used to investigate whether the CES ra tings of healthy community-dwelling speakers were significantly higher than the CES ratings of speakers with dysarthria and PD. To test Hypothesis 2 a one-directional depende nt samples (familial blocking) t-test, = .05 design was used to investigate whether the CE S ratings of 25 particip ants with dysarthria and PD was significantly higher than the CES ratings given to them by their significant others (SO). To test hypothesis 3, a multiple regression analysis was used to determine the relationships between the CES and the tw o measures of speech intelligibility of speakers with dysarthria and PD. The study required recruitment of three groups of subjects. The Parkinson’s disease group designated PD, included indivi duals diagnosed with PD who also had dysarthria. The non-PD group designated NO, included individuals with no existing neurologic injury or disease and no dysart hria, who were matched on gender and meanage to the PD group. The significant others group designated SO, included the significant

PAGE 37

27 others of the individuals in the PD group. Significant other was defined as the person the PD individual identif ied as a primary comm unication partner. Participants We enrolled 25 participants in each of thr ee groups for a total of 75 subjects, based on an a priori power analysis for a one-tailed t-test using = .05, and power = .8. (Faul & Erdfelder, 1996). All participants were community-dwelling volunteers recruited from the University of Florida, Department of Neurology Movement Disorders Clinic, PD support groups, or elder community centers. For all subjects, exclusion criteria incl uded a reported hist ory of neurologic, psychiatric, or language de ficit, Mini-mental State Exam (MMSE) of less than 23 (Folstein, Folstein, & McHugh, 1975), Apathy Scal e less than 14 (Starkstein et al., 1992), non-reader, non-English speaker, unable to foll ow directions due to hearing loss, less than 18 or more than 90 years of age. Indi viduals with PD also had to have a Hoehn &Yahr Staging for Parkinson’s disease (H&Y ) stage between 1 and 4. The H&Y is a staging instrument that is used in PD re search which assesses physical ability. An example of a stage 1 item is symptoms are inconvenient but not di sabling. An example of a stage 5 item is requires constant nur sing care (Hoehn & Yahr, 1967). See Appendix B for all of the items on the scale. The part icipants also had to have a rating between 1 and 4 on the Dysarthria Rating of Severity (D RS). The DRS is a scale that rates the severity of dysarthria from stage 1, no detectable speech disorder, to stage 5, no functional speech (Yorkston et al., 1999). Th e healthy individuals and significant others were excluded if they had a history of a speech impairment. Table 3-1 gives the complete listing of inclusion and exclusion criteria for each group.

PAGE 38

28 Table 3-1. Inclusion/ex clusion criteria for study participants by group Group Inclusion Criteria Exclusion criteria PD n=25 Diagnosis of PD > 6 months Pharmacologically stable Can ID on/off movement cycles Hoehn & Yahr Staged 1-4a MMSE 23b Apathy Scale 14c Dysarthria Functional Limitation Stage 1-4d Functional sentence reading skill Native English speaker Hearing acuity sufficient to follow directions Visual acuity sufficient to read large print Community dwelling 18 to 90 years old History of other neurologic, psychiatric, or language deficits. Hoehn & Yahr staged >4a MMSE < 23b Apathy Scale >14c No dysarthria4 Cannot read Non-English speaker Severely HOH Visually impaired Non-community dwelling No SO, or SO who is unwilling to participate <18 and >90 years old NO n=25 No diagnosed PD MMSE 23b Apathy Scale 14c No Dysarthria Functional sentence reading skill Native English Speaker Hearing acuity sufficient to follow directions Visual acuity sufficient to read large print Community dwelling 18 to 90 years old History of PD, other neurologic, psychiatric, speech, or language deficits MMSE <23b Apathy Scale > 14c Cannot read Non-English speaker Severely HOH Visually impaired Non-community dwelling <18 and >90 years old SO n=25 No history of neurol ogic, psychiatric, speech, or language deficits MMSE 23b Apathy Scale 14c Functional sentence reading skill Native English Speaker Hearing acuity sufficient to follow directions Visual acuity sufficient to read large print Community dwelling 18 to 90 years old History of neurologic, psychiatric, speech, or language deficits MMSE <23b Apathy Scale >14c Cannot read Non-English speaker Severely HOH Visually impaired Non-community dwelling <18 and >90 years old a(Hoehn & Yahr, 1967) b(Folstein et al., 1975), c(Starkstein et al., 1992), d(Yorkston et al., 1999)

PAGE 39

29 Recruitment Parkinson’s Disease Group The Research Coordinator at the UF Movement Disorders Center (UF MDC), Gainesville, FL, who was trained to identify and recruit subjects for clinical trials in PD according to standards established by the NIH, approved a query of the UF MDC Database to identify individuals who met the st udy eligibility criteria to participate in our study (based on IRB-approved Informed Consen t #416-2002). The Principal Investigator (PI) made the initial contact with potential pa rticipants. If the individual indicated a willingness to participate, the PI conducted a telephone screening to ensure that the individual met the inclus ion/exclusion criteria. If the individual met the inclusion/exclusion criteria, an appointment was scheduled to obtain informed consent, perform the screening, and if a ppropriate, perform the study. Non-PD Group Participants were recruited by the PI th rough responses to an IRB-approved printed flyer distributed to communities of older i ndividuals. When a participant responded, the PI conducted a telephone screening to determin e whether or not the participant met the inclusion/exclusion criteria for the study. If the individua l met the inclusion/exclusion criteria an appointment was scheduled to obt ain informed consent, perform the screening, and if appropriate, data collection. Significant Other Group One of the eligibility criteria for the subjects with PD was that they had a significant other who was also wi lling to participate in the study. If the PD subject had a significant other willing to pa rticipate, a tele phone screening was conducted with the

PAGE 40

30 significant other at the same time the PD participant was contacted. Appointments for the PD participant and SO participant were scheduled for the same time. Sample Demographic Data Using the inclusion/exclusion criteria listed in Table 3-1, and the described recruitment procedures, the 75 individuals were recruited in to the study. The demographic data resulted in the follo wing group compositions. The PD group was comprised of 25 individuals with dysarthria and PD, 14 males and 11 females. Their average age was 70.3 years (range 60-84 years, SD = 6.29 years). See Table 3-2 for individual demographics. The non-PD group was made up of 25 individu als, 14 males and 11 females. Their average age was 70.4 years (range 44-79 years, SD = 8.69 years). See Table 3-2 for l demographics of the NO group. The SO Group was made up of 25 individuals 10 males and 15 females. SOs were the spouses of individuals with Parkinson’s disease in 21 of the 25 cases (84%). The other four participants included one sibling, two childre n, and one primary communication partner. See Table 3-2 fo r the demographics for the SO group. Informed Consent The PI obtained informed consent for all seventy-five subjects, using the procedures established and approved in I RB#276-2004. The PI had received all of the required training in the proper procedures fo r obtaining informed consent and protecting the individual’s health care informati on for both HIPAA and NIH compliance.

PAGE 41

31 Screening Measures The inclusion and exclusion cr iteria established to control the threats to the external validity of the study required th at participants complete a sh ort screening battery prior to completion of the study. These measures and the possible threats to validity included: Mini Mental State Examination (MMSE). Cognitive deficits including dementia are typically associated with the overall severity of PD, and may be part of other diseases and aging as well. A short cognitive screening was chosen to control for the unreliable responses that might result from someone with diminish ed cognitive skills in all three groups. The MMSE is widely re cognized as a quick screening tool that gives a gross measure of a person’s cognitive skills (Folstein et al., 1975). We adhered to the MMSE’s esta blished cutoff score of 23 as an indicator of impaired cognition. If the participant obtained a score of 23 or less he/she was excused from further participation in the study. Table 3-2. Demographi c data for all groups PD Group NO Group SO Group Group Participant Age Gender Age Gender Age Gender 01 76 M 66 F 62 F 02 69 F 79 M 59 F 03 69 M 66 F 67 M 04 75 M 77 M 74 F 05 72 M 62 F 64 F 06 68 F 73 M 42 M 07 69 M 72 F 66 F 08 66 F 77 F 67 M 09 63 M 74 F 63 F 10 62 M 73 M 55 F 11 77 F 69 F 75 M 12 69 F 64 F 74 M 13 60 M 68 M 57 F 14 68 M 67 F 66 F 15 73 M 73 M 70 F 16 74 M 73 F 70 F 17 68 F 49 M 70 M 18 65 F 44 M 57 F 19 69 M 79 M 69 F 20 60 F 77 M 62 M 21 84 F 77 M 85 M 22 71 F 77 M 72 M 23 84 M 71 M 75 F 24 69 M 78 M 67 F 25 77 F 75 F 51 M

PAGE 42

32 Apathy Scale. An apathy rating was included because there is evidence that apathy may occur in individuals diagnosed with PD (with and without the presence of depression) (Marin, 1990; Starkstein et al., 1992). Apathy may preclude successful participation in life roles (the domain we intended to investigate in our study). Depression, common to individuals with PD, is routinely recognized and treated in the UF Movement Disorders Center; th erefore a depression screening was not included in this screening battery. Apathy is defined as a lack of motivation that is not attributable to diminished level of consciousness, cognitive impairment, or emotional distress (Marin, 1990; Starkstein et al., 1992). The scale used a cutoff score of >14. See Appendix C for the Apathy Scale. Hoehn and Yahr Staging for Parkinson’s disease (H&Y). We determined that the study would include individuals between H&Y stage 1 and stage 4 to meet the criteria that they be community-dwelling (Hoehn & Yahr, 1967). The H&Y is a five-stage scale designed to rate the disability associ ated with Parkinson’s disease. The stages range from one to five, with one being the mildest (mild unilateral symptoms, inconvenient but not disabling, friends no tice changes in posture, locomotion and facial expression) and five being the most severely disa bled (complete invalidism, constant nursing care required, cannot walk or stand). See Appendix B for the complete scale. Dysarthria Rating of Severity (DRS) We determined that the study would include individuals who had stage 1 dysarthria sever ity (speech problems and increased effort to speak noticeable to the individual), through stage 4 dysarthria severity (severe dysarthria, minimal speech intelligibility and frequent communication breakdowns). The DRS is a 5-stage model that clinicians can use to rank the reported and observed severity of dysarthria of indi viduals with PD (Yorkston et al., 1999). If an individual with PD was observed (and self-reported) to have normal speech, he was excused from participating in the study. Table 3-3 describes the DRS in detail. Table 3-3. Dysarthria Rating of Severity Stage Description Stage 1: No detectable speech disorder Individuals who place high demand on communication will report changes in speech performance Complaints include sense of increased effort to speak Stage 2: Obvious speech disorder with intelligible speech Changes in voice: reduced loudness, reduced pitch variability and unsteady voice Complaints include difficulty speaking in specific communication contexts, su ch as a noisy environment Stage 3: Reduction in speech intelligibility Frequently asked to repeat Imprecise oral articulation Limited range of oral movements Rushes of speech with abnormal placement of pauses

PAGE 43

33 Table 3-3. Continued Stage Description Stage 4: Natural speech supplemented by augmentative techniques Natural speech alone is no longer functional Frequent communication breakdowns Severe dysarthria with minimal speech intelligibility Stage 5: No functional speech Unable to speak at all Requires augmentative and assistive communication devices (Yorkston et al., 1999) Dependent Measures Communication Effectiveness Survey (CES). The CES is the 8-item, 4-point equal interval scale developed in our laborator y. The CES showed initial measurement properties that suggested it w ould be a useful measure of change in communicative effectiveness of individuals w ith the dysarthria of PD. S ee Appendix A for the CES. Sentence Intelligibility (SIT) of the Assessment of In telligibility in Dysarthria Speech (AIDS). The AIDS has standardized instructions for administration, along with established validity, and reliab ility (Yorkston & Beukelman, 1981a, 1981b; Yorkston et al., 1984). It is a measure of connected speech intel ligibility elicited by asking a participant to read a set of sentences aloud. Spontaneous Speech Intelligibility (SSI). Kempler & Van Lanker (2002) showed that a speech intelligibility measure ba sed on a reading sample resulted in significantly higher speech intelligibility scores than scores derived from a spontaneous speech samples of individuals with the dysarthria of PD, although many other investigators have shown that spontane ous speech intelligibility tasks result in significantly higher scores because listener s can derive more from the context of connected speech (Hustad & Beukelman, 2002; Hustad & Beukelman, 2001; Hustad & Cahill, 2003) Because of these conf licting results and because intelligible, spontaneous speech is the ideal outcome of speech therapy we added a measure of spontaneous speech intelligibility to our st udy. Each individual in the PD group was asked to talk for one minute about the pl ace he/she was born after completing the SIT task. Procedures Screening Procedures Once a telephone screening confirmed the indi vidual’s eligibility to participate in the study, an appointment was set up at the si ght of the participant’s choice: home, VA Speech Pathology Clinic, or Shands Speech a nd Hearing Clinic. Informed consent was

PAGE 44

34 obtained prior to the start of every study. It was obtained in the privacy of the individual’s home or in a private treatment room where the participant could ask and answer questions confidentially. After individuals signed the informed consent, the MMSE and Apathy Scale were administered. In addition the individuals diagnosed with PD r eceived the H&Y staging and the DRS rating. The PI administered and scored all screening instruments. Individuals who did not pass the screening were thanked for their time and excused. Individuals who passed the screening continued on to participate in the study. Dependent Measure Collection Procedures The experiment took place in a quiet room of the participant’s choosing. Seventyone participants chose to have the study conducted in their home. The remainder completed the study in the clinic For all participants with PD, the experime nt began 30 minutes after they had taken their medications or were in an “on” medica tion cycle, and ended 30 minutes before they were due to take their medication again or were in an “off” medication condition. We chose to assess participants in the PD gr oup in an “on” medication cycle to obtain optimal speech intelligibility scores to compare with the communicative effectiveness ratings, recognizing that when individuals are in an “o ff” medication cycle speech intelligibility would be decreased. For th e participants without PD the study was scheduled at their convenience. All participants completed the CES i ndependently after re ceiving verbal and written instructions. After the CES was co mpleted every subject was asked to give his/her impressions about the instrument. The PI noted those imp ressions. NO and SO groups were excused after the debriefing.

PAGE 45

35 Following completion of the CES, the PD group completed the SIT by reading 10 sentences that were 14 words in length. Twenty-five unique sets of 10 fourteen-word sentences were developed from the 100 14-wo rd sentences presented in the AIDS to control for listener familiarity (Yorkston et al., 1984). Research Randomizer, a webbased random number generator was used to obtain the 25 unique sets of numbers (Urbaniak & Plous, 1997. http://222randomizer.org). After the sentence inte lligibility task was completed each PD participant generated a one-minute spontaneous speech intelligibility (SSI) sample based on the directions, “I would like you to talk to me for one minut e about the place you were born and grew up. Please keep talking until I tell you to stop.” Scoring Dependent Measures Procedures Three native English speakers who had no hearing loss (based on self-report) transcribed the Sentence Intelligibility (SIT) and Spontaneous Speech Intelligibility (SSI) samples. The transcribers were given both pr inted and verbal instructions from the AIDS for transcribing the samples. Transcribers had no familiarity with or access to the sentence stimuli used. As stated above, sent ences were randomized for each participant to decrease the transcribers’ familiarity with th e sentence intelligibility samples. The PI scored all SIT and SSI transcripts for intelligible and unintelligible words, per AIDS instructions (Yorkston & Beukelman, 1981a). The SIT and SSI score used for statistical analysis was an average of the scores of the three judges. Data Collection Methodology The following methods were used to colle ct, store and transcribe the dependent measures:

PAGE 46

36 Published AIDS manual instructions for te st administration and sample collection were followed (Yorkston et al., 1984). A headset noise-resistant microphone (TS031) positioned 2 cm from the speaker’s mouth captured the speech samples. Microsoft MS Sound Recorder system cont ained in the Dell Inspiron 8600 Laptop Computer was used to record the speech in PM format, at 48.000 kHz, 16 Bit, stereo at 187 kb/sec. Background noise was < 30 dB-A measur ed by a sound level meter throughout recording. The sound level meter was calibrated prior to each recording. Each subject was seated in an upright position, and the microphone was positioned. Each subject received the following instructions in an effort to minimize reading errors: “Please read the sentences to yourself one time. When you are ready, I will ask you to say the number behind the period and then re ad the sentences in your usual speaking voice. Please pause between each sentence. ” Individual samples were saved to the computer hard drive, using a patient iden tification number and date. Recording and transcription parameters were the same as fo r the SIT. The speech samples of the 25 PD individuals were recorded ont o three high fidelity CD’s fo r transcription purposes. The transcribers used “sound” computer software and free field listening in a quiet setting with the listening volume set at a comfortable level to transcribe the samples. Their handwritten responses were returned to the PI for scoring. To ensure that the recordings were of sufficient quality, the three judges we re asked to rate the quality of the speech recording on the CD’s they r eceived as excellent, good, fair, and poor. All three rated the recording quality as good. Inter-rater Reliability Three judges transcribed the sentences and the spontaneous speech samples of the 25 individuals with dysarthria and PD. To a ssess the reliability of three judges’ ratings,

PAGE 47

37 Cronbach’s was used to compare the variation of ratings for a particular judge in relation to the co-varia tion in ratings for all three judges. For intelligible words on the SIT task, the corrected judge to total correlations were .94, .96, and .86 for the three judges respectively yielding a standardized reliability of = .97. For unintelligible words on the SIT task, the corrected judge to to tal correlations were .84, .94, .82 yielding a standardized reliability of = .95. For intelligible words on the SSI task, the corrected judge to total correlations were .98, .96, and .91 for the thr ee judges respectively yielding a standardized reliability of .98. Finally, fo r unintelligible words on the SSI task, the corrected judge to total correlations were .69, .61, a nd .55 for the three judges respectively yielding a standardized reliability of .83. Table 3-6 shows the results of the test for inter-rater reliabilit y. To assess accuracy of scoring, a third party blinded to the study’s purpose transcribed five randomly selected SIT and SSI samples. Accuracy of the transcriptions was 96% agreement be tween the PI and the third party. Table 3-4. Inter-rater reliability Task Judge 1 Judge 2 Judge 3 Standardized Reliability SIT intelligible words .94 .96 .86 =.97 SIT unintelligible words .84 .94 .82 = .95 SSI intelligible words .98 .96 .91 = .98 SSI unintelligible words .69 .61. .55 = .83 Note: SIT=Sentence Intelligibility; SSI=Spontaneous Speech Intelligibility Data Analysis Plan SPSS was used to produce descriptive st atistics and the fo llowing inferential statistical tests: Hypothesis 1: The CES ratings for 25 part icipants with dysarthria and PD will be significantly less than CES ratings of 25 mean genderand age-matched healthy participants. This hypothesis was test ed using a one-tailed dependent samples t-test with the Type 1 error rate set at =.05.

PAGE 48

38 Hypothesis 2: The CES ratings of 25 part icipants with dysarthria and PD will be significantly larger when compared to the CES ratings given to them by their significant others (SO). Due to the dyad ic relationship between the speakers and significant others, this hypothesis was test ed using a one-tailed dependent samples t test with the Type 1 error rate set at = .05. Hypothesis 3: When predicting senten ce intelligibility score (SIT) from communicative effectiveness survey score (CES), spontaneous speech intelligibility score (SSI) accounts for a si gnificant amount of the va riation in SIT above and beyond the variation due to CES for individu als with dysarthria and PD. Likewise, when predicting SSI from CES, SIT accounts for a significant amount of the variation in SSI above and beyond the variation due to CES for individuals with dysarthria and PD. This hypothesis was tested using one-directional multiple regression and comparison of squared semi-partial correlation coefficients. Type 1 error rate set at = .05. The dependent measure was the CES, regressed on SIT and SSI scores.

PAGE 49

39 CHAPTER 4 STUDY RESULTS Hypothesis 1 The CES ratings for 25 participants with dysarthria and PD will be significantly less than CES ratings of 25 genderand mean-age-matched healthy participants. Results of Testing Hypothesis 1 This hypothesis was tested using a one-tailed dependent samples t-test with the Type 1 error rate set at =.05. A paired-samples t-test was used to analyze the mean difference between the PD group and the NO group. The paired-samples test was justified by prior matching of participants between conditions on gender and mean age. The per comparison error rate was controlled at = .05. The correlation between groups was .083 and was not significan tly different from zero ( p = .69). Healthy individuals (NO group) had significan tly higher CES ratings ( M = 28.68, SD = 2.82) than participants with dysarthria and PD ( M = 23.24, SD = 4.82) with a mean difference of 5.44 between groups, t (24) = -4.70, p < .001. The 95% confidence inte rval around the mean difference between groups was 3.05 and 7.83. These data give a measure of construct validity to the CES. See Figure 4-1. Hypothesis 2 : The CES ratings of 25 participants with dysarthria and PD will be significantly higher when compared to th e CES ratings given to them by their significant others (SO). Results of Testing Hypothesis 2 This hypothesis was tested using a one-tailed dependent samples t -test with the Type 1 error rate set at = .05. This analysis compared the mean difference on the CES ratings for individuals with Parkinson’s di sease (PD group) and th e significant others’

PAGE 50

40 (SO Group) CES ratings of the individuals with PD. Although there was a moderate correlation between group ( r = .69), the relationship was not significantly di fferent from zero. The results of the t -test indicate that the PD group participants ha d significantly higher CES ratings ( M = 23.24, SD = 4.82) than their CES ratings given by the significant others (SO group) ( M = 21.64, SD = 5.80) with a mean difference of 1.60, t (24) = 1.864, p = .037. See Figure 4-2. Figure 4-1. Comparison of m eans for PD and NO groups Figure 4-2. Comparison of m eans for PD and SO groups

PAGE 51

41 Hypothesis 3 : When predicting sentence inte lligibility score (SIT) from communicative effectiveness survey score (C ES), spontaneous speech intelligibility score (SSI) accounts for a si gnificant amount of the va riation in SIT above and beyond the variation due to CES for individu als with dysarthria and PD. Likewise, when predicting SSI from CES, SIT accounts for a significant amount of the variation in SSI above and beyond the variation due to CES for individuals with dysarthria and PD. This hypothesis was tested using one-directional multiple regression and comparison of squared semi-partial corre lation coefficients to determine the relationships among the three measures. The dependent measure was the CES, the independent measures were the SIT and SSI scores The multiple regression formula for this hypotheses was: CES = a + bsitXsit + bssiXssi.. Results of Testing Hypothesis 3 Standard multiple linear regression analysis was performed using SPSS REGRESSION to determine the regression co efficients for CES, SIT and SSI. Assumptions were tested by examining normal probability plots of residuals and scatter plots of residuals versus predicted residuals. No violations of nor mality, or linearity were found. SPSS REGRESSION gives th e regression coefficients -intervals, R2, F and pvalues for the regression analysis. Th e full model CES = 2.94 + 47.27(SIT) – 25.46 (SSI) was not significant, F (2, 22) = 1.97, p = 163. R2 = .152, indicated that SIT and SSI accounted for only 15% of the variability in the observations. Table 4-1 gives the statistical data used in the regression model. Table 4-1. Summary of va riables SIT and SSI contri buting to change in CES Unstandardized Coefficients Standardized Coefficients 95% Confidence Interval Model Std. Er. t Sig Lower Bd. Upper Bd. 2.94 16.01 .18 .86 -30.24 36.14 47.27 27.16 .60 1.74 .10 9.06 103.50 (Constant) Avg. % SIT Avg. % SSI -25.45 28.80 .30 .88 .39 -85.18 34.27 Note: CES=Communicative Effectiveness Su rvey; SIT=Sentence Intelligibility; SSI= Spontaneous Speech Intelligibility The partial correlation coefficients ( ) indicated that approximately 15% of the variance of SIT and SSI (11.6% and 3% re spectively) was accounted for in the observations, leaving 85% of the variance unaccounted for. This result could be

PAGE 52

42 attributed to random variability or it could indicate that we did not account for other variables that affected CES ratings. This re sult led to the following post hoc analyses. Post Hoc Analyses In attempting to discover what other va riables contributed to changes in CES ratings, a review of mean CES ratings by H&Y indicated that the PD group rated decreases in CES in relation to increased physical disability and increased dysarthria severity. We felt that post hoc analysis of these severity data was warranted to understand how individuals with dysarthr ia and PD rated their communicative effectiveness in relation to the severity of thei r physical deficits and the severity of their dysarthria. See Figure 4-3. Figure 4-3. CES ratings for H&Y stages of PD We began our post hoc analysis by reviewing our sample characteristics. First, we noted that our sample was skewed toward indi viduals with less seve re physical disability, based on the H&Y ratings (M = 2.08, SD = .996, range 1-4) (where 1 is mild and 5 is total invalidism). Although it was skewed, th e sample met the necessary assumptions for statistical tests used including linearity, and normality (test of kurtosis was not

PAGE 53

43 significant). The same held true for the DRS (M = 2.16, SD = .62, range 1-3) (where 1 is very mild and 5 is no usable speech). See Figure 4-4. Finally, both sentence intelligibility (SIT) and spontaneous speech inte lligibility (SSI) scores neared the ceiling of 100% intelligible (SIT [ M = 94.2%, SD = 6.10%], and SSI [ M = 95.4%, SD = 5.78%]). See Table 4-3 for descriptive statistics. Figure 4-4. CES ratings for ra tings of dysarthria severity Standard multiple linear regression analysis was performed between CES (dependent variable) and the SIT, SSI H&Y, and DRS using SPSS REGRESSION model CES = a + bsitXsit + bssiXssi. + bh&yXh&y + bdrsXdrs. Assumptions were tested by examining normal probability plots of residua ls and scatter plots of residuals versus predicted residuals. No violations of norma lity, or linearity were f ound. Table 4-4gives the complete statistical data for the multiple regression analysis of CES, SIT, SSI, H&Y and DRS.

PAGE 54

44 Table 4-2. Descriptive statistics for CES, SIT, SSI, H&Y & DRS Measure Mean Standard Deviation Number CES Total 23.24 4.82 25 Average %SIT .94 .06 25 Average %SSI .95 .06 25 H&Y 2.08 .99 25 DRS 2.16 .62 25 Note: CES=Communicative Effectiveness Survey; SIT=Sentence Intelligibility; SSI= Spontaneous Speech Intelligibility; H&Y=Hoehn & Yahr Staging of Parkinson’ s disease (Hoehn & Yahr, 1967); DRS=Dysarthria Rating of Se verity (Yorkston et al., 1999) Table 4-3. Summary of variab les SIT, SSI, H&Y, and DRS c ontributing to change in CES Note: CES=Communicative Effectiveness Su rvey; SIT=Sentence Intelligibility; SSI= Spontaneous Speech Intelligibility; H&Y=Hoehn & Yahr Staging of Parkinson’s disease (Hoehn & Yahr, 1967); DRS=Dysarthria Rating of Severity (Yorkston et al., 1999 The full model was sign ificant (F[4, 20] = 6.88, p = .001, R2 = .588), and 59% of the variability in the CES was accounted for by the revised model. Analysis of the standardized coefficients (beta) showed that the H&Y variable was significant ( t = -2.19, p = .040) in this model and accounted for th e largest portion of va riance of all the variables. It was followed by the SSI, which approached significance ( t = -1.08, p =.10). These results suggested that as individuals de clined in physical abil ity, we could predict a decline in their communicative effectiveness (participation co mponent measure) as well, although their mean speech inte lligibility scores (impairment component measure) remained high. This result adds evidence to the content validity of the CES as a measure of communication in soci etal participation. Unstandardized Coefficients Standardized Coefficients 95% Confidence Interval Model B Std. Er. Beta t Sig Lower Bd. Upper Bd. 48.98 15.66 3.13 .005 16.32 81.64 21.04 20.92 .27 1.01 .327 -22.60 64.68 -37.02 27.71 -.44 -1.08 .104 -82.31 8.27 2.72 1.24 -.56 -2.19 .040* 5.30 .13 (Constant) Avg. % SIT Avg. % SSI H&Y DRS 2.13 1.81 -.28 -1.17 .255 5.91 1.66

PAGE 55

45 Participant Feedback After subjects had completed the CES they were asked to respond to three questions, about the survey, and were free to answer or not. Th e questions were: 1) Overall, what is your impression of the CES? 2) Based on your experience, as a person with dysarthria and PD or the significant ot her of a person with dysarthria and PD are there other social speaking situations we shoul d add to this survey? 3) Is there any other information about your communicative effectiv eness that you think other healthcare professionals should know? The PI noted th eir comments on a debriefing form as they were talking. The PI asked for clarifica tion if a comment was unclear. Of the 50 who participated in the study 25 gave comments during the debriefing other than “I thought it was fine” or “I wouldn’t add anything.” See Appendix C for a listing of the comments given by both individuals with dysarthria and PD, and their significant others. In general, participants appreciate the brevit y of the survey. They also reported that the survey tapped most of their daily comm unication situations. Suggestions for added items included: speaking when fatigued, sp eaking when off meds versus on meds, speaking when there is increased noise in th e home, speaking when anxious, rather than angry or upset, speaking when engaged in a physical activity, speaking outdoors, speaking to a group, speaking when on the spot, and speaking for a long period of time. In summary, we were able to accept Hypot hesis 1 and Hypothesis 2. The results of Hypothesis 3 testing were more complex and required post hoc analysis to discern the salient predictors of CES. We were unabl e to accept our hypothesis that SIT and SSI would be significant predictors of change in CES ratings. Post hoc analysis revealed that the only significant predictor of change in CES ratings (based on our model SIT,SSI, H&Y, and DRS) was the H&Y, a measur e of physical disability, followed by

PAGE 56

46 spontaneous speech intelligibility. The contri butions of these results to the field of dysarthria assessment, treatment, and research will be discussed in the next section.

PAGE 57

47 CHAPTER 5 DISCUSSION We began this study by proposing that a measure of communication in the ICF societal participation component was needed for the field of motor speech disorders because although researchers often assume that the treatment effects obtained in the clinic translate to success in an individual’s societal participation to fulfill life roles, they have not often shown it. It is impossible to investigate the phenomenon of communication in societal participation at this time beca use a psychometrically sound measure of communication in the societal participation component does not yet. We proposed that the Communicative Effectiveness Survey, a survey which measures communicative effectiveness in 8 everyday speaking s ituations, may serve as a measure of communication in societal participation, based on the litera ture, and our early foundational work in analyzing the measurement properties of the CES The purpose of this study was to investig ate three types of va lidity, face validity, content validity, and construc t validity, of the CES for i ndividuals with dysarthria and PD. The results of this study, their contribu tions to the dysarthria literature, and their implications for the future direction of research needed will be discussed first. The study limitations will be discussed next. We w ill end the discussion by describing ways to improve the CES, based on input from the study participants. Hypothesis 1 McHorney described different types of validity that must be addressed in survey development (McHorney et al., 2002). Some of those types of valid ity include ecological

PAGE 58

48 validity, social validity, construc t validity, and clinical validit y, what others call construct validity (Cook & Campbell, 1979; Maxwell & Delane y 2004; Shadish, Cook, & Campbell, 2002). We presented the results of two analyses that showed foundational evidence for the content validity and face validity of the CES for individuals with PD and a wide range of speech competence in Chapter 2. The purpose of Hypothesis 1 was to investigate another aspect of validity, construct validity (specifically known-group construct validity), which is th e ability of a measure to disc riminate between groups with and without disease. The result of the test of Hypothesis 1 showed that CES ratings were significantly higher for healt hy, community-dwelling individu als (NO group) than they were for community-dwelling individuals with dysarthria and Park inson’s disease (PD group) matched for gender and mean age. There are many other components of cons truct validity that remain to be investigated for the CES. Questions that re quire further investigat ion include whether or not the CES ratings are sensitive to: 1) di fferences among individuals with PD with different levels of functiona l ability; 2) differences am ong individuals with PD with different levels of dysarthria severity; 3) di fferences between indivi duals with dysarthria and PD and individuals with PD only, 4) differences between CES ratings when individuals with PD are in “ on” and “off” medication periods, 3) differences between preand post-treatment, and 4) differences th at occur as the disease progresses. Although the NO group (healthy, community-d welling individuals) rated their CES significantly higher than th e PD group, they did not judge their communicative effectiveness very effective 100% of the time ( M = 28.68, SD = 2.82). Based on this finding, we propose that there are factors other than speech that contribute to judgments

PAGE 59

49 about communicative effectiveness. Several factors mentioned by participants included: feeling ill at ease when speaking with a stra nger or at a social event, feeling selfconsciousness when speaking, and the heari ng acuity of communication partners. Determining what factors, other than speec h, contribute to communicative effectiveness is an area that requires further investigation. Hypothesis 2 It has been documented that individuals w ith PD underestimate their difficulties, or are not aware of the severity of their deficits (Marsde n, Parkes, & Quinn, 1981; Yorkston et al., 1994; Yorkston et al ., 2004). Reasons for these misperceptions have been hypothesized to occur because of dementia (Bodis-Wollner, 2003) and frontal executive dysfunction (Ho et al., 2002). However there is an emerging literatur e that suggests that individuals with PD have sensorimotor defici ts that contribute to a mismatch between their actual performance that their judgment of performa nce (Abbruzzese & Berardelli, 2003; Bodis-Wollner, 2003; Ho, Bradshaw, Ians ek, & Alfredson, 1999). One example of this deficit pertinent to the LSVT (discussed ea rlier) is that individuals with PD perceive the loudness of their voice differently th an age-matched individuals without PD (Abbruzzese & Berardelli, 2003). There is also literature that showed that caregivers of indi viduals with chronic diseases rated those individuals significantly more disabled than the individuals rated themselves (Andresen, Vahle, & Lollar, 2001; Doyle et al., 2004; Duncan et al., 2002; Segal & Schall, 1994; Sneeuw, Aaronson, deHaan, & Limburg, 1997). Hypothesis 2 was designed to investigate the question of the differences between the ratings of communicative effectiveness of individuals wi th dysarthria and PD and the CES ratings that their significant others gave them. The results of th e test of Hypothesis 2 showed

PAGE 60

50 that individuals with dysarthria and PD rated their communicative effectiveness significantly higher than the si gnificant others rated them Our results add to the evidence in the respondent literature that i ndividuals with PD rate their performance higher than do their significant others. Ho wever, with regard to the motor speech literature, our results from Hypothesis 2 differed from those of Ball and colleagues (2004), who reported that individuals with ALS and their caregivers did not have significantly different ratings on the CETI-M (a survey very similar to the CES). We suggest that the difference in our results lends further evidence to the literature that found individuals with PD do not (or unable to) recognize thei r communication deficits. Our results indicated that obtaining info rmation from the individual with the communication disorder alone might not give researchers and clinicians the most accurate information about the individual’s co mmunicative effectiveness. Our evidence holds clinical importance for training the communicative dyad, rather than the individual with dysarthria and PD alone. This is consistent with the find ings of others in the field who work with individuals with PD and motor speech disorders (Hustad, 1999; Hustad, Beukelman, & Yorkston, 1998; Yorkston et al., 1994; Yorkston et al., 2004). If the goal of treatment is improved communicative effectiveness for improved participation in life roles, and there is a si gnificant difference betw een the communicative dyad’s ratings of communicative effectiveness (i.e., for the in dividual with dysarthria and PD) then training the communica tive dyad should begin at the earliest stage of treatment rather than waiting until close to discharge. Such training would provide the clinician, patient, and significant other an opportunity to try to reach agreement about what has occurred during treatment and what more need s to be done. However, we need evidence

PAGE 61

51 to demonstrate the benefit of such a treatment paradigm. A future research project could evaluate the differences in CES ratings betw een individuals with dysarthria and PD and significant others who received treatment of the dyad, compared to a group where the individual with dysarthria and PD received treatment individually and the significant other received the traditional fa mily education information. Our research was not designed to determ ine whether the difference found between ratings of the PD group and the SO group was due to decreased insight, increased frontal dysfunction, or the proxy effect. In estab lishing our inclusion/ex clusion criteria we attempted to control for depression (client s from the MDC are routinely treated for depression), dementia (MMSE), and apathy (Apathy Scale). However, we did not address the possibility that ot her sorts of frontal executive dysfunction could be present. An area for future investigation would be to determine what role frontal dysfunction (if it is present) contributes to th e CES ratings of individuals wi th dysarthria and PD. This kind of research would be an example of wh at the World Health Organization envisioned for rehabilitation research when conceptu alizing the ICF (WHO, 2001). The WHO ICF recommends research that spans the entire model, body function and structure (identifying frontal dysfunction) activity limitations (identifying the deficits that result from frontal dysfunction), and participati on restrictions (identifying the difficulty completing life roles because of the frontal dysfunction). Hypothesis 3 The speech intelligibility literature has documented a number of threats to the internal validity of a study th at uses speech intelligibility as a dependent variable. Such threats include listener familiarization (Hus tad & Cahill, 2003; Spitzer et al., 2000), length of utterance (Spitzer et al., 2000), sp eaking task (Kempler & Van Lancker, 2002),

PAGE 62

52 medication fluctuations (Goberman & Blomgre n, 2003; Goberman et al., 2002; Larson et al., 1994; Mawdsley & Gamsu, 1971; Quaglieri & Celesia, 1977), practice (Inzana et al., 1996) and attention (Ho et al ., 2002). We attempted to co ntrol for these threats when we developed Hypotheses 3, by stipulating how our speech samples would be obtained, who would transcribe them, what tasks woul d be used, when the speech samples would be collected (i.e., when the individual was in the “on” state of medi cations). In addition, we stipulated the level of practice we w ould let the speakers do before recording the speech samples. Finally we attempted to cont rol for attentional deficits by conducting the study in the quietest room of an individual’s home. However, the re sults of Hypothesis 3 were not as we had hypothesized. Neither SI T nor SSI accounted for a significant portion of the variance in the CES in our initial multiple regression analysis. Review of the data led us to postulate that the results of the multiple regression analysis were not significant because the va riability within the SI T and SSI scores was small (individuals in the study had high SIT and SSI scores) although the data met the statistical assumptions required for regressi on analysis. Of note, the SIT approached significance in the initial multiple regression mode l (SIT and SSI only). It is possible that with a larger sample SIT would have proven significant for predicting change in CES. We also found that in our sample, the two speech intelligibility measures (SIT and SSI) were redundant in the multiple regression m odels. Further research will be done to determine which variable was the best predictor of ch ange in CES. In the post hoc analysis, using a multiple regression model that accounted for the two speech intelligibility measures (SIT a nd SSI), a measure of physical disability (H&Y), and dysarthria severity (DRS), the variable that accounted for the largest portion

PAGE 63

53 of variability of observations in CES was the H&Y ( p = .04), a measure of physical disability, followed by the SSI (whi ch approached significance at p= .10). The results of this post hoc analysis gave us evidence that the CES was associated with a measure of participation (the H&Y) more than the m easures of impairment (SIT or SIT). We stated in the introducti on that there is not a one-to-one relationship between speech intelligibility scores and successf ul functional communication (Weismer & Laures, 2002; Yorkston, 1996). Our results supp orted this literature. High SIT and SSI scores did not predict how an individual rated his communicative effectiveness in everyday speaking situations. Rather, ba sed on our results, the rating of physical disability taken from the H&Y was the only significant predictor of change in CES ratings. We recognized at the time of data anal ysis that our PD group was made up of individuals who had less phys ical impairment and dysarthria severity based on H&Y ratings and DRS ratings. We also recognized that the PD group had SIT and SSI scores that were relatively high ( M = .94 and M = .95 respectively). In the future studies should be designed so that each level of severity is equally represented, and the numbers of participants in each group are large enough to ensure the power of our results are adequate to ensure the statis tical rigor of our results. The most important implication of the fi nding of Hypothesis 3 wa s that individuals with dysarthria and PD rate d their communicative effectiveness less effective on everyday speaking situations although their sp eech intelligibility scores were high. We propose that this result demonstrated that the CES was tapping a different component of communication than the one tapped by either of the sentence intelligibility scores. We

PAGE 64

54 suggest that this is evidence that the SIT a nd the SSI are measures of activity (as they have been presented to be) and the CES is a measure of communication in the societal participation component of the ICF. Some might argue that an individuals with a speech intelligibility score of 95% should not be a candi date for speech therapy. However, if an individual was able to dem onstrate decreased communicative effectiveness on the CES, that kept him from fulfilling his life roles in societal participation, he might well be a candidate for speech therapy. We are in the ea rly states of survey development trying to determine the validity and reliability of the CE S; however another area that remains to be studied in the future is that of the sensitivity and specificity of the CES to identify those individuals whose communicative effectiveness is limiting their participation in fulfilling their life roles. Limitations of the Study This study was not without limitations. The most obvious limitation, based on our post hoc analysis was that there are other f actors that need to be taken into account besides an individual’s speech when attemptin g to measure communicative effectiveness, such as physical ability. Additionally, we received input from both healthy communitydwelling individuals and community-dwelling in dividuals with PD and their SOs that other factors affected their judgments of how they spoke, including being ill-at-ease in speaking situations, being self-conscious about PD, being uncomfortable around strangers, being put on the spot, and having c onversational partners who were hard of hearing, and/ or impatient. Next, our SIT and SSI measures were redundant variables in the multiple regression models we devised. In the fi rst multiple regression analysis model SIT

PAGE 65

55 approached significance ( p = .10) and in the second (when severity was added into the model) the SSI approached significance ( p = .10). In addition to the problem of redundancy, th e literature presents conflicting reports on what role a spontaneous speech sample plays in determining speech intelligibility versus the traditional intelligibility measur e where words and sentences are read by the speaker. Some authors report improved intell igibility beause the listener an derive meaning from the context of what is being said (Dagenais, Garcia, & Watts, 1998; Dagenais, Watts, Turnage, & Kennedy, 1999; Hustad & Beukelman, 2002; Hustad & Beukelman, 2001; Hustad & Cahill, 2003). Othe rs have found that spontaneous speech samples result in decreased intelligibility ra tings (Kempler & Van Lancker, 2002). In our sample, the SIT and SSI tasks were redundant, which could lead one to suggest choosing just one speech intelligibility task. However, we also reviewed the transcription methodology we chose to determine if it woul d point to methodological problems we had overlooked. A review of the inter-ra ter reliability on the SSI, although still good, was 10%-15% less than that of the SIT. Refer to Ta ble 3-6. We used the AIDS transcription instructions for both intelligibility tasks (Y orkston et al., 1984). A key of correct sentences was available to check against the transcribed SIT sample s. However, there were no keys with which to compare the SSI transcriptions. In the future, if a spontaneous speech sample is used, the PI c ould transcribe the SSI samples immediately after they are collected while the content of the sample wa s fresh. That transcription could serve as the key with which to comp are the transcribed samples. This method

PAGE 66

56 might reduce the amount of discrepancy between the intelligible words and unintelligible words in the SSI samples. Improving the Communicative Effectiveness Survey Survey development typically involves ge tting input from individuals with the problem being studied and th eir significant others through focus groups or individual interviews. Such input gives ecological/social validit y to a survey (Ant onius et al., 1995; Lomas et al., 1989; Yorkston et al., 1994; Yorkston et al., 2001), which in turn should increase the likelihood of content validity a nd face validity of an instrument. Lomas and colleagues’ (1989) designed the Communicativ e Effectiveness Index (CETI) to measure changes in the functional communication of indi viduals with aphasia. They used input from individuals with aphasia and caregive rs during CETI development. Yorkston and colleagues (1999) modified the CETI to measure changes in the functional communication of individuals with motor speech disorders (such as the dysarthria associated with PD). As with the Loma s group, Yorkston and colleagues (1999) gathered input from individuals with a variety of motor speech disorders before modifying the CETI and called it the CES (the measure used in our study). Because individuals with motor speech di sorders had already given input into the survey, we did not use focus groups ahead of time but rather asked for feedback from the individuals with dysarth ria and PD and their significant others after they had completed the CES. There were only a few addition items suggested. In the future, we will consider adding items to the instrument based on the i nput from the participants. Once items are added, further analysis would be made to ensure the stability of the measurement properties of the CES.

PAGE 67

57 We believe the information participants gave in response to the third debriefing question “is there any other information about your communicative e ffectiveness that you think other healthcare professi onals should know?” clearly show ed that individuals with PD and their SO’s viewed communicative e ffectiveness as a societal participation measure. Examples of their responses include d: feeling very self -conscious about one’s speech, having trouble communicating when bei ng put on the spot, feeling tense made speech worse, being perceived as cognitively impaired because it is difficult to respond quickly. See Appendix C for a full listing of comments to all questions. The groups’ responses also encompassed mo re than communication. Other things they thought healthcare professionals should kno w included: fear about what the future holds, disease progression, abil ity to remain independent, the strain the disease puts on marriage and family, personal safety, financia l security, physical a nd emotional ups and downs, role reversals in relati onships, stigma of PD (tremors, reduced mobility, using a wheelchair, slowness of thinking and res ponding), and impatience of the public. Many of the concerns expresse d had little to do with speech per se, but had much to do with maintenance of relationships and i ndependence, both of which require certain levels of communicative effectiveness. Uncovering the impact that decreased communicative effectiveness has on those livi ng with PD (both patient and significant others) is an area that is ripe for further re search. This kind of research (qualitative research) is emerging in the aphasiology lit erature (Damico & Simmons-Mackie, 2003; Kagan & LeBlanc, 2002; Lyon, 2000), but there is very little of it being systematically done in the area of motor speech disorders in general, and hypokinetic dysarthria of Parkinson’s disease in particular (Antoni us et al., 1995; Yorkst on et al., 1994).

PAGE 68

58 Future research questions that are of part icular interest relative to doing qualitative research in the area of communicative effectiv eness of individuals wi th dysarthria and PD include: 1) what does the phr ase communicative eff ectiveness mean to you? 2) what are the biggest societal barrier s you face to communicate effect ively? 3) how do medication fluctuations affect your communicative effectivenes s? 3) do you manage your medications to maximize communicative effectiven ess in societal part icipation and if so, how? 4) do you manage your medications for othe r kinds of societal participation, and if so, how? Answers to these questions would be beneficial in further establishing the validity of communicative effectiveness as a measure of societal participation. Conclusion In conclusion, this study has provided ev idence that adds to the face validity, content validity, and construct validity of the CES as a measure of communication for the societal participation compone nt the WHO ICF model. Th e CES differentiated healthy, community-dwelling elders from gendera nd mean-age-matched individuals with dysarthria and PD. In add ition, we showed that there was a significant difference between the way individuals with dysart hria and PD rated their communicative effectiveness and the way their significant others rated the communicative effectiveness of the individual with PD. Finally, we f ound evidence that indivi duals with dysarthria and PD rated their communicati on less effective, as their physical ability declined although their sentence intelligibility scores a nd spontaneous speech in telligibility scores remained high. In motor speech disorders, we have ma ny measures of body f unctions/structures (perceptual judgments, acoustic, and phys iologic measures) and activity (speech intelligibility), but we have yet to devel op a measure of communication in societal

PAGE 69

59 participation. Such a measure is needed by speech therapists to demonstrate that the treatments they administer at the clinical le vel result in success for the client where the client needs it, to fulfill life roles at home, at work in the community. Such measures are also needed by researchers interested in moto r speech disorders, so that as we go about designing studies to demonstrate the efficacy and effectiveness of new treatments we remember the goal of speech therapy is to enable a person to communicate more effectively to participate in his life roles. Although there are many questions left to be answered regarding the reliability and other types of validity of the CES for individuals with dysar thria and PD, we believe we have presented evidence that the CES holds promise as a measure of communication for societal participation to fulfill life roles.

PAGE 70

APPENDIX A COMMUNICATIVE EFFECTIVENESS SURVEY

PAGE 71

61 NAME: ____________________________ DATE: ______________________ In this survey we ask you to rate how effective your speech is in different communication situations. Please read each stat ement. Then rate how effectively you communicate in that situation. If you feel your speech is very effective, mark the 4. If your speech does not allow you to communicate at al l in a situation, mark the 1. Feel free to use any number on the scale. 1. Having a conversation with a family member or friends at home. Not able to do Very Effective 1 2 3 4 2. Participating in conversation w ith strangers in a quiet place. Not able to do Very Effective 1 2 3 4 3. Conversing with a familiar person over the telephone. Not able to do Very Effective 1 2 3 4 4. Conversing with a stra nger over the telephone. Not able to do Very Effective 1 2 3 4 5. Being part of a conversation in a no isy environment (social gathering). Not able to do Very Effective 1 2 3 4 6. Speaking to a friend when you are em otionally upset or you are angry. Not able to do Very Effective 1 2 3 4 7. Having a conversation while traveling in a car. Not able to do Very Effective 1 2 3 4 8. Having a conversation with someone at a distance (across a room). Not able to do Very Effective 1 2 3 4 Hustad, K.C., 1999. Optimizing Communicative Effectiveness: Bringing it together. In Yorkston, K.M., Beukelman, D.R., Strand, E.A, & Bell, K.R. (Eds.), Management of Motor Speech Disorders (2nd ed.), (pp.438-541).

PAGE 72

62 APPENDIX B HOEHN AND YAHR STAGING FOR PARKINSON’S DISEASE Name:___________________________________ Date: _______________ 1. Stage One 1. Signs and symptoms on one side only 2. Symptoms mild 3. Symptoms inconvenient but not disabling 4. Usually present with tremor of one limb 5. Friends have noticed changes in postu re, locomotion and facial expression 2. Stage Two 1. Symptoms are bilateral 2. Minimal disability 3. Posture and gait affected 3. Stage Three 1. Significant slowing of body movements 2. Early impairment of equilib rium on walking or standing 3. Generalized dysfunction that is moderately severe 4. Stage Four 1. Severe symptoms 2. Can still walk to a limited extent 3. Rigidity and bradykinesia 4. No longer able to live alone 5. Tremor may be less than earlier stages 5. Stage Five 1. Cachectic stage 2. Invalidism complete 3. Cannot stand or walk 4. Requires constant nursing care Hoehn, M.M. & Yahr, M.D. (1967). Parkins onism: onset, progression, and mortality. Neurology 17(5): 427-442.

PAGE 73

APPENDIX C PARTICIPANT FEEDBACK ON THE CES

PAGE 74

64 Table C-1. Participant feedback Question Individuals with Dysarthria and PDSignificant Others 1. Overall, what is your impression of the survey? -Succinct, clear -Good and short -Short is a good thing -I like short tests -Short & sweet -You need more numbers. I would have liked to put some in-between -Simple and understandable -Is this enough questions to be meaningful? -I am neutral on it -It was quick and easy -I’m benign on it -Good and short -I may be a 3.5 on some -Some of the wording is not clear -It doesn’t really apply to most of our problems with lifestyle changes -Very lifelike and applicable 2. Are there other social speaking situations we should add to this survey? -Perhaps a question about conversing for a lengthy period of time -I don’t see anything about my speech when I am tired -Outdoors is a problem, I can’t yell like I used to -Hard to order food at a restaurant with the noise and the rush -Maybe something about talking faster-that is my wife’s complaint -Most of my friends are hard of hearing. I think that is the problem. -I don’t answer the phone anymore so that question doesn’t really apply to me. -Add something about fatigue level and speaking. -You need a question about talking when you are in an anxious situation, or being judged, not just mad or whatever. -Change the emotionally upset question. I don’t let myself get mad. Maybe use the term emotionally involved. -It is hard to talk and dance and we love to dance. -I worry when he is outdoors by himself he can’t yell at me for help. -Maybe ask about talking at home with the TV on or when there are other people here—it makes a difference. -When all the grandchildren are here I can’t hear him. Maybe one about talking when he is tired. -When more than two of us are here at home it is hard to hear him and he cannot seem to get into the conversation—it is as though we pass him by.

PAGE 75

65 Table C-1 continued. Question Individuals with Dysarthria and PDSignificant Others 3. Is there any other information about your communicative effectiveness that you think other healthcare professionals should know? -I notice my voice is quieter, but I don’t think it makes much difference, does it? -I am newly diagnosed but worry about the future and what is going to happen. -Nobody realizes how much I fluctuate from day to day, not because of meds, but just how I feel in general -I have trouble speaking when I am put on the spot for an answer -People seem to think I am more cognitively impaired because of my slowness to respond Everyone notices how clear my speech is when I am mad -Everything is harder now -Exercise helps my fatigue -My voice is worse when I get confused, like when I am tired. -I want everyone to know that therapy helps. I think people interact differently with me because I come across as a competent communicator now and I speak up. -Feeling tense makes my speech worse -The hearing of the listener is very important -I have become very selfconscious about my speech -My husband won’t go and leave me alone although I want him to because I can take care of myself very well. -I don’t want to be a burden to my wife, but I already am -It is hard on the person I am with. I have lost my facial expression and he can’t tell when I am happy. -Everything takes a long time to do and people aren’t always patient. -No one really addresses caregiver burden. I am trapped here with him. -He knows everyone around here and is embarrassed by some of his problems. No one has talked to us about that kind of thing. -PD is very hard on a marriage and nobody tells you that. -The fear of what is coming is almost unbearable. -I feel desperate. -I take Prozac to manage my feelings of dread for what is ahead. -We are both getting older and our kids are up north. We don’t have much help. -I have to do all kinds of things I never thought I would do. -I don’t like making decisions by myself, but sometimes I just do. -It takes us so long to do anything that it is easier to stay home.

PAGE 76

66 LIST OF REFERENCES Abbruzzese, G., & Berardelli, A. (2003) Sensorimotor integration in movement disorders. Movement Disorders, 18 (3), 231-240. Adams, S. G. (1997). Hypokinetic dysarthria in Parkinson's disease. In M. R. McNeil (Ed.), Clinical Management of Sensorimotor Speech Disorders (pp. 2261-2285). New York: Thieme. Andresen, E., Vahle, V., & Lollar, D. (2001). Proxy reliability: heal th-related quality of life measures for people with disability. Quality of Life Research, 10 (7), 609-619. Ansel, B. M., & Kent, R. D. (1992). Acoustic-p honetic contrasts and intelligibility in the dysarthria associated with mixed cerebral palsy. Journal of Speech and Hearing Research, 35 296-308. Antonius, K., Beukelman, D. R., & Reid, R. (1995). Communication disability of Parkinson's disease. In D. A. Robin, K. M. Yorkston & D. R. Beukelman (Eds.), Disorders of Motor Speech: recent adv ances in assessment, treatment, and clinical characterization (pp. 2275-2286). Baltimor e: Paul H. Brooks. Ball, L. J., Beukelman, D. R., & Pattee, G. L. (2004). Communication effectiveness of individuals with amyotr ophic lateral sclerosis. Journal of Communication Disorders, 37 197-215. Bergner, M., Bobbitt, R. A., Carter, W. B ., & Gilson, B. S. (1981). The sickness impact profile: development and final revisi on of a health status measure. Medical Care, 19 787-805. Beukelman, D. R., & Yorkston, K. M. (1977). A communication system for the severely dysarthric speaker with an intact language system. Journal of Speech and Hearing Disorders, 42 265-270. Beukelman, D. R., & Yorkston, K. M. (1979) The relationship between information transfer and speech intelligibi lity of dysarthric speakers. Journal of Communication Disorders, 12 189-196. Bodis-Wollner, I. (2003). Neuropsychologican and perceptual def ects in Parkinson's disease. Parkinsonism and Related Disorders, 9 S83-S89. Bond, T. G., & Fox, C. M. (2001). Applying the Rasch Model: Fundamental measurement in the human sciences Mahwah, NJ: Lawrence Erlbaum Associates.

PAGE 77

67 Brandt, E. N., Jr., & Pope, A. M. (1997). Models of disability and re habilitation. In E. N. Brandt, Jr. & A. M. Pope (Eds.), Enabling America: assessing the role of rehabilitation scienc e and engineering (pp. 24-39). Washington, D.C.: National Academy Press. Cardol, M., deJong, B. A., van den Bos, G., Be elen, A., deGroot, I. J. M., & deHaan, R. J. (2002). Beyond disability: perceived participation in people with a chronic disabling condition. Clinical Rehabilitation, 16 27-35. Cook, T.D. & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis for field settings. Chicago: Rand McNally. Dagenais, P. A., Garcia, J. M., & Watts, C. R. (1998). Acceptability and intelligibility of mildly impaired dysarthric speech by diffe rent listeners. In M. P. Cannito, K. Yorkston & D. Beukelman (Eds.), Neuromotor Speech Disorders: nature, assessment, and management (pp. 213-227). Baltimore : Paul H. Brookes. Dagenais, P. A., Watts, C. R., Turnage, L. M., & Kennedy, S. (1999). Intelligibility and acceptability of moderately dysarthri c speech by three types of listeners. Journal of Medical Speech-Language Pathology, 7 91-96. Damico, J. S., & Simmons-Mackie, N. N. (2003). Qualitative research and speechlanguage pathology: A tutorial for the clinical realm. American Journal of Speech-Language Pathology, 12 131-143. Darley, F. L., Aronson, A. E., & Brown, J. R. (1975). Motor Speech Disorders Philadelphia: W.B. Saunders. Doyle, P., McNeil, M. R., JM, M., Prieto, L., Hula, W., Lustig, A., Ross, K., Wambaugh, J., Rothi, L., & Elman, R. (2004). The Bu rden of Stroke Scale (BOSS) provides valid and reliable score estimates of functioning and well-being in stroke survivors with and without communication disorders. Journal of Clinical Epidemiology, 57 (10), 997-1007. Duffy, J. R. (1995). Motor Speech Disorders: substrates, differential diagnosis, and management St. Louis: Mosby. Duffy, J. R. (2005). Motor Speech Disorders: Substrates, differential diagnosis, and management (2nd ed.). St. Louis, MO: Elsevier. Duncan, P., Lai, S., Tyler, D., Perera, S., Reker, D., & Studenski, S. (2002). Evaluation of proxy responses to the stroke impact scale. Stroke, 33 (11), 2593-2599. Enderby, P. (2000). Assessment and treatment of functional communication in dysarthria. In L. E. Worrall & C. M. Frattali (Eds.), Neurogenic Communication Disorders: a functional approach (pp. 247-261). New York: Thieme.

PAGE 78

68 Faul, F., & Erdfelder, E. (1996). G*Powe r: a general power analysis program. Behavior Research Methods, Instruments, & Computers, 28 1-11. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state. A practical method for grading the cognitive st ate of patients for the clinician. Journal of Psychiatric Research, 12 189-198. Gilson, B. S., Gilson, J. S., Bergner, M., Bobb itt, R. A., Kressel, S., Pollard, W. E., & Vesselago, M. (1975). The sickness impact profile development of an outcome measure of health care. American Journal of Public Health, 65 1304-1310. Goberman, A. M., & Blomgren, M. (2003). Park insonian speech disf luencies: effects of l-dopa-related fluctuations. Journal of Fluency Disorders, 28 (1), 55-70. Goberman, A. M., Coelho, C., & Robb, M. (2002). Phonatory characteristics of Parkinsonian speech before and afte r morning medicati on: the ON and OFF states. Journal of Communication Disorders, 35 217-239. Hammen, V. L., & Yorkston, K. (1996). Sp eech and pause characteristics following speech rate reduction in hypokinetic dysarthria. Journal of Communication Disorders, 29 429-445. Ho, A. K., Bradshaw, J. L., Iansek, R., & Alfredson, R. (1999). Speech volume regulation in Parkinson's disease: effects of im plicit cues and explicit instructions. Neuropsychologia, 37 1453-1460. Ho, A. K., Iansek, R., & Bradshaw, J. L. (2002). The effect of a concurrent task on parkinsonian speech. Journal of Clinical and Expe rimental Neuropsychology, 24 (1), 36-47. Hoehn, M. M., & Yahr, M. D. (1967). Parkin sonism: onset, progression and mortality. Neurology, 17 (5), 427-442. Hustad, K. C. (1999). Optimizing communicative effectiveness: bringing it together. In K. Yorkston, D. Beukelman, E. A. Strand & K. R. Bell (Eds.), Management of Motor Speech Disorders in Children and Adults (pp. 483-537). Austin, TX: Proed. Hustad, K. C., & Beukelman, D. (2002). List ener comprehension of severely dysarthric speech: effects of linguistic cues and stimulus cohesion. Journal of Speech and Hearing Research, 45 545-558. Hustad, K. C., & Beukelman, D. R. (2001). Effects of linguistic cues and stimulus cohesion on intelligibility of severely dysarthric speech. Journal of Speech and Hearing Research, 44 497-512. Hustad, K. C., Beukelman, D. R., & Yo rkston, K. M. (1998). Functional outcome assessment in dysarthria. Seminars in Speech and Language, 19 (3), 291-302.

PAGE 79

69 Hustad, K. C., & Cahill, M. A. (2003). E ffects of presentation mode and repeated familiarization on intelligibility of dysarthric speech. American Journal of Speech-Language Pathology, 12 198-206. Inzana, C., Driskell, J., Salas, E., & Johnson, J. (1996). Effects of preparatory information on enhancing performance under stress. Journal of Applied Psychology, 81 429-435. Jette, A. M. (2003). Assessing disab ility in studies on physical activity. American Journal of Preventative Medicine, 25 (3), 122-128. Jette, A. M., Haley, S. M., & Kooyoomjian, J. T. (2003). Are the ICF activity and participation dimensions distinct? Journal of Rehabilitation Medicine, 35 (3), 145149. Kagan, A., & LeBlanc, K. (2002). Motivati ng for infrastructure change: toward a communicatively accessible, participation-ba sed stroke care system for all those affected by aphasia. Journal of Communication Disorders, 35 153-169. Kempler, D., & Van Lancker, D. (2002). Effect of speech task on intelligibility in dysarthria: a case study of Parkinson's disease. Brain and Language, 80 449-464. Kent, R. D. (2000). Research on speech motor control and its disorders: a review and prospective. Journal of Communication Disorders, 33 (5), 391-428. Kent, R. D., & Read, C. (1992). The Acoustic Analysis of Speech San Diego: Singular Publishing Group. Kent, R. D., Vorperian, H. K., Kent, J. F ., & Duffy, J. R. (2003). Voice dysfunction in dysarthria: application of the Multi-Dimensional Voice Program(tm). Journal of Communication Disorders, 36 281-306. Keysor, J. J., & Jette, A. M. (2001). Have we oversold the be nefit of latelife exercise? Journal of Gerontology: Medical Sciences, 56A (7), M412-M423. Klasner, E. R., Yorkston, K. M., & Strand, E. A. (1999). Patters of perceptual features in speakers with ALS: a preliminary study of prominence a nd intelligibility considerations. Journal of Medical Speech-Language Pathology, 7 (2), 117-125. Larson, K., Ramig, L., & Scherer, R. (1994). Acoustic and glottographic voice analysis during drug-related fluctuati ons in Parkinson disease. Journal of Medical SpeechLanguage Pathology, 2 227-239. Linacre, J. M. (1991). Winsteps Ministep Rasch-model computer programs [computer software]. Linacre, J. M. (1994). Sample size and item calibration stability. Rasch Measurement Transactions, 7 (4), 328.

PAGE 80

70 Linacre, J. M. (2002). Optimizing ra ting scale category effectiveness. Journal of Applied Measurement, 3 (1), 85-106. Lomas, J., Pickard, L., Bester, S., Elbard, H., Finlayson, A., & Zoghaib, C. (1989). The communicative effectiveness index: devel opment and psychometric evaluation of a functional communication measure for adult aphasia. Journal of Speech and Hearing Disorders, 54 (1), 113-124. Lyon, J. G. (2000). Finding, defining, and refini ng functionality in real life for people confronting aphasia. In L. E. Worrall & C. M. Frattali (Eds.), Neurogenic Communication Disorders: a functional approach (pp. 137-161). NY: Thieme. Marin, R. S. (1990). Differential diag nosis and classification of apathy. American Journal of Psychiatry, 147 22-30. Marsden, C. D., Parkes, J., & Quinn, N. (1981). Fluctuations of disabi lity in Parkinson's disease-clinical aspects. In C. D. Marsden & S. Fahn (Eds.), Movement Disorders (pp. 96-122). London: Butterworth. Mawdsley, C., & Gamsu, C. (1971). Peri odicity of speech in Parkinsonism. Nature, 231 315-316. Maxwell, S.E. & Delaney, H.D. (2004). Designing Experiments and Analyzing Data (2nd ed). Mahwah, NJ: Lawrence Erlbaum Associates. McHorney, C. A., Robbins, J., Lomax, K., Rosenbek, J. C., Chignell, K. A., & Kramer, A. E. (2002). The SWAL-QOL and SWAL-CARE outcomes tool for oropharyngeal dysphagia in adults: III. Do cumentation of reliability and validity. Dysphagia, 17 97-114. Nordenfelt, L. (2003). Action theory, disability and ICF. Disability and Rehabilitation, 25 (18), 1075-1079. Parkinson's Disease Soci ety website (2005). URL: http://www.parkinsons.org.uk Last accessed 04/16/05. Perenboom, R. J. M., & Chorus, A. M. J. (2003). Measuring participation according to the ICF. Disability and Rehabilitation, 25 (11-12), 577-587. Pollard, B., & Johnston, M. (2001). Problem s with the sickness impact profile: a theoretically based analysis and a pro posal for a new method of implementation and scoring. Social Science and Medicine, 52 921-934. Quaglieri, C., & Celesia, G. (1977). Effect of thalamotomy and levodopa therapy on the speech of Parkinson patients. European Journal of Neurology, 15 34-39.

PAGE 81

71 Ramig, L. O. (1994). Voice treatment for patien ts with Parkinson's disease: development of an approach and preliminary efficacy data. Journal of Medical SpeechLanguage Pathology, 2 (3), 191-209. Ramig, L. O. (1995). Comparison of two forms of intensive speech treatment for parkinson disease. Ramig, L. O., Countryman, S., O'Brien, C., Hoehn, M., & Thompson, L. (1996). Intensive speech treatment for patients with Parkinson's disease: shortand longterm comparison of two techniques. Neurology, 47 (6), 1496-1504. Ramig, L. O., Countryman, S., Thompson, L., & Horii, Y. (1993). Comparison of two forms of intensive speech treatment for Parkinson disease. Journal of Speech and Hearing Research, 38 (5), 1232-1251. Ramig, L. O., Countryman, S., Thompson, L. L., & Horii, Y. (1995). Comparison of two forms of intensive speech treatment for Parkinson disease. Journal of Speech and Hearing Research, 38 (6), 1232-1235. Ramig, L. O., Sapir, S., Countryman, S ., Pawlas, A., O'Brien, C., Hoehn, M., & Thompson, L. L. (2001). Intensive vice treatment (LSVT) for patients with Parkinson's disease: a 2 year follow up. Journal of Neurology, Neurosurgery and Psychiatry, 71 493-498. Rasch, G. (1980). Probabilistic models for some in telligence and attainment tests (Expanded) Chicago: University of Chicago Press. Rosenbek, J. C., & LaPoint, L. L. (1978). The dysarthrias: descri ption, diagnosis and treatment. In F. L. Darley (Ed.), Clinical Management of Neurogenic Communication Disorders Boston: Little, Brown and Company. Segal, M., & Schall, R. (1994). Determining functional/health status and its relation to disabillity in stroke survivors. Stroke, 25 2391-2397. Shadish, W.R., Cook, T.D., & Campbell, D.T. (2002). Experimental and QuasiExperimental Designs for Ge neralized Causal Inference Boston: HoughtonMifflin. Sneeuw, K., Aaronson, N., deHaan, R. J., & Limburg, M. (1997). Assessing quality of life after stroke: the value and limitations of proxy ratings. Stroke, 28 1541-1549. Spitzer, S. M., Liss, J. M., Caviness, J. N ., & Adler, C. (2000). An exploration of familiarization effects in the percepti on of hypokinetic and ataxic dysarthric speech. Journal of Medical Speech-Language Pathology, 8 285-293.

PAGE 82

72 Starkstein, S. E., Mayberg, H. S., Preziosi, T. J., Andrezejewski, P., Leiguarda, R., & Robinson, R. G. (1992). Reliability, validity, and clinical correlates of apathy in Parkinson's disease. Journal of Neuropsychiatry and Clinical Neurosciences, 4 134-139. Urbaniak, G. C., & Plous, S. (1997). Research Randomizer 2004, from http://222.randomizer.org Van Den Eeden, S. K., Tanner, C. M., Bernst ein, A. L., Fross, R. D., Leimpeter, A., Bloch, D. A., & Nelson, L. M. (2003). Incide nce of Parkinson's disease: variation by age, gender, and race/ethnicity. American Journal of Epidemiology, 157 (11), 1015. Velozo, C. A., & Peterson, E. W. (2001). Deve loping meaningful fear of falling measures for community dwelling elderly. American Journal of Physical Medicine and Rehabilitation, 80 (9), 662-673. Weismer, G., & Laures, J. S. (2002). Direct magnitude estim ates of speech intelligibility in dysarthria: effects of a chosen standard. Journal of Speech and Hearing Research, 45 (3), 421-433. WHO. (2001). International Classification of Func tioning, Disability and Health: ICF Geneva: WHO. Wright, B. D. (1997). A history of social science measurement. Educational Measurement: Issues and Practice, 16 (4), 33-45. Wright, B. D., & Linacre, J. M. (1989). Ob servations are always ordinal; measurements, however, must be interval. Archives of Physical Me dicine and Rehabilitation, 70 (12), 857-860. Wright, B. D., & Stone, M. H. (1979). Best Test Design: Rasch Measurement Chicago, IL: MESA Press. Yorkston, K. M. (1996). Treat ment efficacy: dysarthria. Journal of Speech and Hearing Research, 39 S46-S57. Yorkston, K. M., & Beukelman, D. R. (1981a). Assessment of intelligibility of dysarthric speech. In. Tigard: CC Publications. Yorkston, K. M., & Beukelman, D. R. (1981b) Communication efficiency of dysarthric speakers as measured by sentence intelligibility and speaking rate. Journal of Speech and Hearing Disorders, 46 (3), 296-301. Yorkston, K. M., Beukelman, D. R., St rand, E. A., & Bell, K. R. (1999). Management of Motor Speech Disorders in children and Adults (2nd ed.). Austin, TX: Pro-ed.

PAGE 83

73 Yorkston, K. M., Beukelman, D. R., & Tice, R. (1996). Speech In telligibility Test [speech analysis] (Version 1.0). Lincoln, NE: Tice Technology Services. Yorkston, K. M., Beukelman, D. R., & Traynor, C. D. (1984). Assessment of Intelligibility of Dysarthric Speech Austin, TX: Pro-ed. Yorkston, K. M., Bombardier, C., & Hamme n, V. L. (1994). Dysarthria from the viewpoint of individuals with dysarthria. In J. Till, K. Yorkston & D. Beukelman (Eds.), Motor Speech Disorders: advances in assessment and treatment Yorkston, K. M., Dowden, P. A., & Beukelman, D. R. (1992). Intelligibility measurement as a tool in the clinical management of dysarthric speakers. In R. D. Kent (Ed.), Intelligibility in Speech Disorders: theory, measurement and management (pp. 2265-2286). Amsterdam: John Benjamins. Yorkston, K. M., Klasner, E. R., & Swans on, K. M. (2001). Communi cation in context: a qualitative study of the experiences of individuals with multiple sclerosis. American Journal of Speech-Language Pathology, 5 126-137. Yorkston, K. M., Miller, R. M., & Strand, E. A. (2004). Management of Speech and Swallowing in Degenerative Diseases (2nd ed.). Austin, TX: Pro-ed. Yorkston, K. M., Strand, E. A., & Kennedy, M. R. T. (1996). Comprehensibility of dysarthric speech: implications for assessment and treatment planning. American Journal of Speech-Language Pathology, 5 55-66.

PAGE 84

74 BIOGRAPHICAL SKETCH Neila J. Donovan received her Bachelor of General Studies degree with a major in speech pathology and minors in education and audiology from the University of Kansas, Lawrence, Kansas in January 1983. She receiv ed her Master of Arts degree in speechlanguage pathology in July 1985, also from th e University of Kansas. She currently holds the Certificate of C linical Competency from the American Speech-LanguageHearing Association. She plans to remains ac tive in both clinical practice and clinical research to improve the communicative effec tiveness of individuals with acquired motor speech disorders after graduation.


Permanent Link: http://ufdc.ufl.edu/UFE0010055/00001

Material Information

Title: Extending Dysarthria Research with a Measure of Communicative Effectiveness
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010055:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010055/00001

Material Information

Title: Extending Dysarthria Research with a Measure of Communicative Effectiveness
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010055:00001


This item has the following downloads:


Full Text












EXTENDING DYSARTHRIA RESEARCH
WITH A MEASURE OF COMMUNICATIVE EFFECTIVENESS

















By

NEILA J. DONOVAN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
2005

































To those who taught me to love learning















ACKNOWLEDGMENTS

I wish to thank my mentor Jay Rosenbek (counselor, teacher, master clinician, and

friend) for his tireless efforts in all of these roles. I am grateful to him for seeing what I

have to offer and for encouraging me to set my own direction. I appreciate him for being

patient when he could have been otherwise. He leads by example; watching him I

understand what it means to be both clinician and scientist of the highest caliber.

I thank my committee members (Craig Velozo, Mary Ellen Young, and Diane

Kendall) for giving me their time and the benefit of their expertise. Their insights have

been rewarding and their encouragement gratifying. I believe this committee should be

recognized as an example of the ideal committee. They remained committed to my

doctoral education and progress from the first meeting to the dissertation defense. They

always convened in the spirit of making good science through intellectual exchange and

mutual respect. They helped shape me into a budding scientist who values rigorous and

sometimes challenging scientific discussions among colleagues.

I thank the people who helped me complete this dissertation in a timely fashion.

They include Nan Musson, who served as my VA liaison and co-principal investigator at

the Malcom Randall VA. She smoothed the way for me at every turn. I thank the

individuals who volunteered to transcribe the speech samples, Elizabeth, Summer,

Jennifer, and Jessica. I thank Michael Okun, Kelly Foote and Hubert Fernandez, the

physicians at the UF Movement Disorders Center who allowed me to have access to their

patients during both survey development and dissertation data collection. Their entire









team is outstanding, but I want to thank Sarah Munson and Chuck Jacobsen in particular

for all their help in gleaning the database. Finally, although I cannot mention them by

name I thank all of the individuals who volunteered to participate in this study. They had

to open their homes to me and give of their time and effort. Everyone was cordial and

hospitable. I will treasure the personal stories they shared with me.

I could not have made it to this point in my life without the love and support of

family and friends. I cannot begin to name everyone who has helped me get to this point

in my life and to the culmination of this fantastic experience called graduate education,

for fear of leaving someone out. I was most fortunate to get good genes and solid,

Midwestern nurturing that gave me a love of reading, a steadfast work ethic, and the

desire to help others. I have been doubly blessed to find mentors and friends at every

turn on my educational and professional path. I wish to thank them all. Finally I wish to

acknowledge my husband Michael Hammer, who has always encouraged me to follow

my dreams.















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ......... .................................................................................... iii

LIST OF TABLES ............................... .............. ..... ......... .. vii

LIST OF FIGURES .............................................. ........ ...... ............. viii

ABSTRACT ........ .............. ............. ...... ...................... ix

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Statem ent of th e P rob lem ..................................................................... .. ............
Sp ecific A im s ..................................................... ..................... 4
R research H ypotheses .................. .......................................... ........ .. ..5

2 LITER A TU R E REV IEW ............................................................. ....................... 6

World Health Organization International Classification of Functioning and
D disability ........................................... ............... ............... ... .. .. ........... .6
Assessment and Treatment of the Dysarthria of Parkinson's Disease (PD)..............
M easurement in Dysarthria......................... ......... ........................... 16
Operationalizing Communicative Effectiveness .............. .......................................18
Development of the Communicative Effectiveness Survey (CES)............................21

3 DESIGN AN D M ETH OD S............................................... ............................. 26

P a rtic ip a n ts ........................................................................................................... 2 7
R ecru itm ent ................................................................................... ..................... 2 9
Parkinson's D disease G roup............................................ ........... ............... 29
N on -P D G rou p ............ .......................................................................... .. ...... .. 2 9
Significant O their G roup ........................................................... .....................29
Sam ple D em graphic D ata ............................................................ ............... 30
Inform ed Consent .................................... .. .... ..... .. ............30
Screening M measures .................. ................................ ...... .. .......... .... 31
D dependent M measures ..................................................... ... .. ........ .... 33
P ro c e d u re s .......................................................................................3 3
Screening Procedures .............................................. ........ ................ ......33
Dependent M measure Collection Procedures.............................. ............... 34









Scoring Dependent M measures Procedures ................................. ............... 35
D ata C collection M ethodology.......................................................... ............... 35
Inter-rater R liability ............................................. .... .. .. ................36
D ata A n aly sis P lan ............ ........................................................................... .. ....... .. 3 7

4 ST U D Y R E SU L T S .................................................................... ... ........................ 39

R results of Testing H ypothesis 1 ........................................ ........................... 39
R results of Testing H ypothesis 2 ........................................ ........................... 39
R results of Testing H ypothesis 3 ........................................ ........................... 41
Post Hoc Analyses ................................. ........ ... ........ ......... 42
Participant Feedback ................................................. ........ .............. ... 45

5 DISCUSSION .................. ................................... ........... ............... 47

H y p o th e sis 1 ............................................................................................................... 4 7
H y p o th e sis 2 ............................................................................................................... 4 9
Hypothesis 3 ............................. ........... ...................51
Improving the Communicative Effectiveness Survey .................................... 56
C conclusion ...................................................................................................... ....... 58

APPENDIX

A COMMUNICATIVE EFFECTIVENESS SURVEY ..............................................60

B HOEHN AND YAHR STAGING FOR PARKINSON'S DISEASE ........................62

C PARTICIPANT FEEDBACK ON THE CES ................................................... 63

L IST O F R E F E R E N C E S ............................................................................................. 66

B IO G R A PH IC A L SK E T C H ....................................................................................... 74
















LIST OF TABLES


Table page

2-1 SIP communication and social interaction items ........................................... 15

2-2 Demographics of participants in two Rasch analyses of CES ratings.................23

2-3 Results of two Rasch analyses of CES ratings................................. ... ..................24

2-4 Summary of measured steps 4-point rating scale PD sample.............................25

2-5 Summary of CES 4-point Rating Scale Characteristics of PD Sample....................25

3-1 Inclusion/exclusion criteria for study participants by group...............................28

3-2 Demographic data for all groups.................................................................. 31

3-3 D ysarthria R eating of Severity........................................................ ............... 32

3-4 Inter-rater reliability ................................................ ............... 37

4-1 Summary of variables SIT and SSI contributing to change in CES.........................41

4-2 Descriptive statistics for CES, SIT, SSI, H&Y & DRS .......................................44

4-3 Summary of variables SIT, SSI, H&Y, and DRS contributing to change in CES...44

C -l P participant feedb ack ........................................................... ................ ................64
















LIST OF FIGURES


Figure p

4-1 Comparison of means for PD and NO groups......... ..........................................40

4-2 Comparison of means for PD and SO groups ..... ......... ................................40

4-3 CES ratings for H&Y stages of PD ... ........................................................ ................ 42

4-4 CES ratings for ratings of dysarthria severity .............................................. .43















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EXTENDING DYSARTHRIA RESEARCH
WITH A MEASURE OF COMMUNICATIVE EFFECTIVENESS
By

Neila J. Donovan

May 2005

Chair: John C. Rosenbek
Major Department: Communicative Disorders

Researchers often assume but have not shown that treatment effects obtained in

clinic translate to success in an individual's societal participation. Testing this

assumption in the field of motor speech disorders is impossible at present because a

psychometrically sound measure of communication in societal participation does not

exist. The best the profession can do is assess speech intelligibility, a valid and reliable

measure of connected speech used to demonstrate change at the clinical level.

The Communicative Effectiveness Survey (CES) may be an instrument to extend

dysarthria research into societal participation, however, as yet it has unknown

psychometric properties.

We asked three experimental questions to study the face validity and construct

validity of the CES:

* Does the CES differentiate individuals with dysarthria and Parkinson's disease
(PD) from non-PD gender and mean-age-matched individuals?









* Do individuals with dysarthria and PD rate their CES higher than their significant
others (SO) rate them?

* Do traditional measures of sentence intelligibility (SIT) and spontaneous speech
intelligibility (SSI) predict changes in CES?

A convenience sample of 75 participants was divided into three groups of 25:

individuals with dysarthria and (PD);non-PD group; and SOs of individuals with PD.

The dependent variables were the CES (all groups), the SIT and SSI (PD group only).

Results showed that the non-PD group rated their CES significantly higher than did

the PD group. The PD group rated their CES significantly higher than the SO group

rated them. The multiple linear regression model for SIT and SSI as predictors of change

in CES was not significant (F[2, 22] = 1.97,p = .163). The model accounted for 15% of

the variability (R2 = .152). A post hoc multiple linear regression analysis, adding

measures of physical disability (Hoehn and Yahr [H&Y]) and dysarthria severity (DRS)

was significant (F[4,20] = 6.88, p = .001), and 59% of the variability was accounted for

(R2= .588). H&Y was the largest (3 = .599) and only significant predictor of CES ratings

(t = -2.19, p = .040).

The results of this study added evidence to the face validity and the construct

validity of the CES. However, to ensure that the CES is a valid and reliable measure of

communication in societal participation more work needs to be done on the other types of

validity and reliability.














CHAPTER 1
INTRODUCTION

Statement of the Problem

Researchers often assume but have not shown that the treatment effects obtained in

the clinic translate to success in an individual's societal participation to fulfill life roles.

Testing this assumption in the field of motor speech disorders is impossible at present

because a psychometrically sound measure of communication in societal participation

does not exist. The best the profession can do is assess speech intelligibility (a measure

of connected speech), which has demonstrated its worth as a measure of change at the

clinical level. Or we can measure perceptual, acoustic and physiologic changes in

speech. However, if the goal of speech therapy is to enable a person to communicate

more effectively to participate in his life roles (what some call functional

communication), then measures of communication in societal participation are needed.

One such measure that is under development is a measure of communicative

effectiveness.

Models of rehabilitation such as the World Health Organization (WHO)

International Classification of Functioning, Disability, and Health (ICF), provide us with

a rationale for extending treatment (such as speech therapy) and rehabilitation research

into the societal participation component. Conceptually we propose that extending

research into the societal participation component of the ICF model requires a different

measure called communicative effectiveness. We define communicative effectiveness as

a person's ability to actively and efficiently get his message across successfully in his









home and community settings to fulfill life roles (Hustad, 1999; Lomas et al., 1989;

Yorkston, Klasner, & Swanson, 2001).

Neurologic diseases such as Parkinson's disease often cause impairments including

a class of speech disorders called the dysarthrias. The typical clinical and research

approach to the dysarthrias is to measure and treat the signs of impairment, which can

include sound imprecision, disturbed rate, reduced loudness and pitch variability.

Dysarthria results in problems in verbal communication (speech) "due to paralysis,

weakness, abnormal tone, or incoordination of the speech musculature" (Duffy, 2005). A

principal measure of dysarthria treatment outcome is speech intelligibility, a measure

based on the acoustic presentation of a series of increasingly complex speech productions

that a listener can decode, from single words to complex sentences (Yorkston, 1996;

Yorkston & Beukelman, 1981a, 1981b; Yorkston, Beukelman, & Traynor, 1984).

The tacit assumption of this approach is that changes in speech intelligibility will

have equally positive influences on a person's communication to fulfill life roles; what

the World Health Organization (WHO) International Classification of Functioning,

Disability, and Health (ICF) identifies as the component of participation (WHO, 2001).

However, studies report no clear one-to-one relationship between a speech intelligibility

score and a person's communication success beyond the clinical treatment setting

(Weismer & Laures, 2002; Yorkston, 1996).

As the population ages we expect the incidence of PD to increase worldwide (Van

Den Eeden et al., 2003). It has been estimated that approximately 70% of individuals

with PD eventually develop speech disorders that affect not only them but their families

as well. (Parkinson's disease Society, 2004) Therefore it is important that we investigate









aspects of communication disorders that are pertinent to this group at the level where the

difficulties are the most troublesome, in fulfilling their life roles-the ICF's participation

domain.

The Communicative Effectiveness Survey (CES) is being studied in our laboratory

to serve as a measure of communication in the ICF societal participation component. The

CES originated from an aphasiology questionnaire, the Communicative Effectiveness

Index, developed by Lomas and colleagues (1989). Ball and colleagues (2004) reported

on the psychometric properties of an instrument very similar to the CES (CETI-M) that

they used to measure the communicative effectiveness of 25 individuals with

amyotrophic lateral sclerosis (ALS) (one type of movement disorder that often results in

motor impairments and dysarthria). Those authors demonstrated the usefulness of their

communicative effectiveness measure (CETI-M) to identify social communication needs

of persons with ALS and their significant others, people for whom traditional speech

therapy is often contra-indicated (Ball et al., 2004). They also reported that the CETI-M

showed psychometric properties including face validity and content validity for what

today is classified as the participation component of the ICF. However, because the

speech impairments of individuals with ALS make up only one subset of the dysarthrias,

research is needed that includes individuals with other types of dysarthria as well. We

chose to investigate one of those groups, individuals with hypokinetic dysarthria and

Parkinson's disease.

In a prior study we investigated the measurement properties of the Communicative

Effectiveness Survey (CES) using a single-parameter model of item response theory

called Rasch analysis. We conducted two analyses: first for 95 individuals with various









movement disorders and speech competency from normal to severe dysarthria; and

second for 66 individuals with idiopathic PD and speech competency from normal to

severe dysarthria. Rasch analysis gives information that can be compared to the

traditional test-theory terms such as validity, reliability and responsiveness. Our results

indicated that the CES demonstrated sufficient measurement properties including: the

theoretical hierarchy matched the sample hierarchy of items (similar to face validity),

individuals with movement disorders and arrange of speech competence endorsed the

items on the survey in a predictable manner (similar to content validity, and there was

high (.90) person-to measure correlation (comparable to Cronbach's alpha), similar to

traditional measures of reliability of response (for both groups). The results were

considered strong enough to prompt us to continue investigating its usefulness as a tool

that we could use to measure Communication at the level of societal participation. We

used the data from the earlier analyses to design this study to add further validity to the

instrument for individuals with PD.

The goal of this study was designed to add evidence to the face validity, content

validity and construct validity of the Communicative Effectiveness Survey for individuals

with dysarthria and PD by accomplishing three specific aims.

Specific Aims

* Specific aim 1: Determine if CES ratings differentiate normal speakers from
dysarthric speakers with PD.

* Specific aim 2: Determine the difference in CES ratings by participants with
dysarthria and PD and their CE ratings completed by significant others.

* Specific aim 3: Determine the relationship between two measures of speech
intelligibility and CES ratings of participants with dysarthria and PD.









Research Hypotheses

Based on the established specific aims, we developed three experimental

hypotheses.

* Hypothesis 1: The CES ratings for 25 participants with dysarthria and PD will be
significantly lower than CES ratings of 25 mean gender- and age-matched healthy
participants.

* Hypothesis 2: The CES ratings of 25 participants with dysarthria and PD will be
significantly higher than the CES ratings given to them by their significant others
(SO).

* Hypothesis 3: When predicting sentence intelligibility score (SIT) from
communicative effectiveness survey score (CES), spontaneous speech intelligibility
score (SSI) accounts for a significant amount of the variation in SIT above and
beyond the variation due to CES for individuals with dysarthria and PD. Likewise,
when predicting SSI from CES, SIT accounts for a significant amount of the variation
in SSI above and beyond the variation due to CES for individuals with dysarthria and
PD.













CHAPTER 2
LITERATURE REVIEW

To understand the development of our thesis that dysarthria research needs to be

extended into the ICF societal participation component, we divided review of the

literature into four sections which will include: a description of the components of World

Health Organization's ICF and how they pertain to research and treatment of dysarthria; a

brief summary of the state of the art in assessment and treatment of hypokinetic

dysarthria (the dysarthria that results from Parkinson's disease); a review of the

traditional treatment outcome measures of dysarthria; and an operational definition of

communicative effectiveness; and a review of how the CES has been tested in other

laboratories. The final section of the chapter gives the foundational evidence that the

CES had sufficient measurement properties to continue investigating its validity as a

measure of communication in societal participation for individuals with idiopathic PD

and dysarthria.

World Health Organization International Classification of Functioning and
Disability

The WHO International Classification of Functioning, Disability and Health (ICF)

is one of the models used in the rehabilitation literature (WHO, 2001). The ICF is a

compelling conceptual model for two reasons. First, the ICF's biopsychosocial design

attempts to integrate the most important aspects of both the medical model and the social

model of functioning and disability. The ICF defines health-state as a bi-directional

dynamic process involving the person with the disease/illness, which can be positively or

negatively affected by contextual/ environmental and personal factors. It is now









recognized that a constellation of environmental and personal factors may positively or

negatively impact health-state and functioning, regardless of the type or intensity of

treatment received (Cardol et al., 2002; Jette, 2003; Keysor & Jette, 2001). Second, the

ICF is a compelling model because it illustrates the complexities involved in planning

and carrying out investigations and interventions to achieve and maintain optimum

functioning of individuals in compromised health states (Brandt & Pope, 1997).

The ICF includes 2 components: body structures and functions; and activity and

participation (previously know as the domains of impairment, handicap, and disability)

(WHO, 2001). Activity is defined as a discrete task or action that a person executes.

Participation is defined as a complex task that may require multiple sets of skills to

complete. Chapter 3 of the ICF describes the activity and participation skills needed for

communication. A communication activity for speech production is "producing words,

phrases and longer passages in spoken messages with literal and implied meaning, such

as expressing a fact or telling a story in oral language" (p. 134) (WHO, 2001).

Communication for societal participation is defined as "holding a conversation: starting

and sustaining an interchange of thoughts and ideas, carried out by means of spoken,

written, sign or other forms of language, with one or more people one knows or who are

strangers in formal or causal settings" (p. 135) (WHO, 2001). Impairment is the term

used to denote a deficit in body structures and/ or body functions (i.e. paralysis of a body

as a result of cerebral infarct). Activity limitations and participation restrictions are the

terms used when an individual cannot successfully complete the delineated skills under

the conditions defined by the ICF.









The inclusion of these domains, allows for the possibility of research to differentiate

between an individual's capacity to perform an activity and his actual participation in

fulfilling societal roles. However, there are now published reports by investigators that

many different types of raters including healthcare providers and rehabilitation scientists

have difficulty differentiating between activity limitation and participation restriction

(Jette, Haley, & Kooyoomjian, 2003; Nordenfelt, 2003; Perenboom & Chorus, 2003).

These reports provide initial evidence that the theoretical distinction between activity and

participation may be more understandable than the concrete application of rating them in

rehabilitation research, assessment, and treatment.

The science of rehabilitation is moving toward an integrated approach that includes

consideration of performance, context of performance, and psychosocial issues thanks to

conceptual models of enablement such as the ICF. Unfortunately, most studies of

dysarthria have not kept up with the current advances in rehabilitation (Enderby, 2000).

Rather, most studies of dysarthria are based on perceptual judgments, acoustic, and

physiologic measures. Speech-language pathologists make perceptual judgments during

an individual's speech evaluation. Perceptual judgments are typically made using a

rating scale to judge the voice quality, pitch, loudness, prosody, intonation, hypernasality,

and precision of articulation of speech. Acoustic measures include fundamental

frequency, vocal intensity, maximum duration of phonation, semitone standard deviation.

Physiologic measures include things such as forced vital capacity, maximum inhalatory

pressure, and maximum exhalatory pressure (Ansel & Kent, 1992; Hammen & Yorkston,

1996; Kent, 2000; Kent & Read, 1992; Kent, Vorperian, Kent, & Duffy, 2003; Klasner,

Yorkston, & Strand, 1999; Rosenbek & LaPoint, 1978). These kinds of measures are









related most directly to the body structures/functions component of the ICF (Enderby,

2000). Specifically, the ICF Body Structures and Functions Chapter 3 "Voice and

Speech Structures/Voice and Speech Functions" describes the body structures and

functions required to produce voice and speech.

On the other hand, the commonly used assessment of connected speech production,

the Assessment of Intelligibility in Dysarthric Speech (AIDS), would be classified in the

activity component (Ball et al., 2004; Yorkston & Beukelman, 1981a, 1981b; Yorkston,

Beukelman, & Tice, 1996; Yorkston et al., 1984; Yorkston, Dowden, & Beukelman,

1992; Yorkston, Strand, & Kennedy, 1996). For example the ICF Activity/Participation

Chapter 3 states that communication addresses receiving, and producing different forms

of communication including words, gestures and signs, pictures, writing, and body

language, along with using augmentative communication devices.

Although the ICF provides the framework for a more integrated rehabilitation

approach in dysarthria research, such research has been hindered by an absence of valid

and reliable tools for measuring communication beyond the body structure/function and

activity components of the ICF. Pertaining to the hypokinetic dysarthria associated with

PD, the bulk of the literature would be classified in the body structures/functions

component of the ICF (WHO, 2001).

There is scant documentation of what impact, if any, hypokinetic dysarthria has on

an individual's participation in fulfilling his societal roles. However, according to

Yorkston and colleagues (2001), "there remains an urgent need to develop measures that

reflect the quality of communication participation in adults with acquired neurologic

communication disorders" (p.136).









We do have an instrument that was developed to show change in communicative

effectiveness of individuals with aphasia before and after treatment. That instrument is

the Communicative Effectiveness Index (CETI). We suggest that the CETI has d content

validity as a measure of participation. The Communicative Effectiveness Index (CETI)

items were developed from interviews that involved individuals with aphasia and their

caregivers about the everyday speaking situations where individuals experienced

decreased communicative effectiveness (Lomas et al., 1989). Additionally, the CETI was

designed to show changes in communicative interactions in social situations for

individuals with aphasia. The authors reported that individuals with aphasia

demonstrated improved communicative effectiveness after therapy based on changes in

ratings of communicativeness (Lomas et al., 1989). Of late the psychometric properties

of the CETI-M (an instrument very similar to the CES) were reported based on the

ratings of individuals with dysarthria and amyotrophic lateral sclerosis (Ball et al., 2004).

These authors also presented evidence of the face validity of the CETI-M as a measure of

communication in societal participation.

In conclusion, we suggest that the WHO ICF and other conceptually driven models

of functioning and disability provide the rationale for an instrument that will measure an

individual's communication to successfully participate in life roles. We do not have such

an instrument at this time in the field of speech pathology. We suggest that the CES may

be an instrument that was originally designed to be such a measure and that there is early

evidence that it has usefulness for individuals with amyotrophic lateral sclerosis. We do

not, however, know how individuals with PD would rate their communicative

effectiveness on the CES.









Assessment and Treatment of the Dysarthria of Parkinson's Disease (PD)

Any neurologic disease afflicting the bulbar system can result in a motor speech

problem called dysarthria. Parkinson's disease (PD) is such a disease and the resulting

motor speech problem is called hypokinetic dysarthria. Parkinson's disease can affect

500,000 people at any one time (Van Den Eeden et al., 2003). An estimated 70% of

individuals diagnosed with PD experience dysarthria ranging from mild to severe over

the course of the disease (Yorkston, Miller, & Strand, 2004). Dysarthria can affect the

person as well as the speech. Consider this statement presented on the Parkinson's

disease Society website: speech disorders:

can interfere with all aspects of communication and as a result there can be a great
deal of unhappiness, frustration, and misunderstanding, not only for the person with
PD but also for the significant other and the people the person comes into contact
with (Parkinson's disease Society, 2005 http://www.parkinsons.org.uk).

Dysarthria is not a single speech disorder, but several different speech disorders. Duffy

(2005) defined the dysarthrias broadly as:

the collective name for a group of speech disorders resulting from disturbances in
muscular control over the speech mechanism due to damage of the central or
peripheral nervous system. It designates problems in oral communication due to
paralysis, weakness, abnormal tone, or incoordination of the speech musculature
(p. 5).

Specifically, the hypokinetic dysarthria associated with PD has been identified

based on the following perceptual characteristics: reduced intensity (monoloudness),

reduced stress and intonation patterns (monotone, monopitch), abnormal voice quality,

articulatory imprecision of consonants produced in rushed bursts of speech, and

inappropriate or illogical pauses during speech (Adams, 1997; Darley, Aronson, &

Brown, 1975; Duffy, 1995, 2005; Rosenbek & LaPoint, 1978; Yorkston et al., 2004).









Ramig and colleagues built a substantial body of literature showing the immediate

and long term treatment efficacy of the Lee Silverman Voice Treatment (LSVT) for

individuals with hypokinetic dysarthria and PD (Ramig, 1994, 1995; Ramig,

Countryman, O'Brien, Hoehn, & Thompson, 1996; Ramig, Countryman, Thompson, &

Horii, 1993; Ramig, Countryman, Thompson, & Horii, 1995; Ramig et al., 2001). Lee

Silverman Voice Treatment is an intensive therapy (4 hours per week for 16 weeks and

daily home practice) based on the theory that training maximum phonatory effort leads to

increased vocal intensity that, in turn, generalizes to other speaking situations.

Ramig has reported that changes in one or more acoustic or physiologic measure

(fundamental frequency, vocal intensity, maximum duration of phonation, semitone

standard deviation, and forced vital capacity) and perceptual ratings (loudness, pitch,

monotonicity) were significantly higher for individuals who received LSVT compared to

those who received a treatment of respiratory effort at post-treatment, 12 months post-

treatment, and 24 months post-treatment intervals (Ramig, 1994, 1995).

If we look at how the authors categorized their assessment tools with regard to the

ICF components of body structures/functions and activity/participation, we see that self-

ratings and family-ratings of the perceptual categories of loudness, monotonicity, and

hoarseness were described as measures of functional communication. Ratings of speech

intelligibility and conversational initiation were described as measures of activity

limitations (Ramig, 1995). Finally, they categorized ratings on the Communication and

Social Interaction subtests of the Sickness Impact Profile (SIP) as measures of functional

outcome (Ramig et al., 1996; Ramig et al., 1995). The SIP (Bergner, Bobbitt, Carter, &

Gilson, 1981; Gilson et al., 1975) is defined by its authors as a behaviorally-based









measure of sickness-related dysfunction, rather than a functional outcomes measure

(Pollard & Johnston, 2001).

Ramig group's use of pre- and post-treatment self-report ratings as functional

communication outcome measures bears review. Results of these self-report ratings for

males and females in both treatment groups resulted in changes that were statistically

significant for all perceptual measures (except loudness) including: monotonicity,

hoarseness, intelligibility, and initiation of conversation. Self-ratings of loudness resulted

in a time, treatment, gender interaction whereby males who received LSVT showed the

greatest improvement. These results showed that from the patients' perspectives, both

treatments resulted in improvement on all variables rated (except loudness) for a specific

group of males and a specific group of females. On face value, these results added

credence to the theorized variability of self-report ratings from individuals PD. For

example, other authors reported that individuals with mild to moderate dysarthria and PD

had reduced insight into their speech deficits (Antonius, Beukelman, & Reid, 1995;

Yorkston, Bombardier, & Hammen, 1994).

Other considerations, however, may have to do with study methodology. The

authors state that two speech-language pathologists administered treatment to all

individuals and "worked closely together to ensure consistency and equivalent high effort

and motivation across both forms of treatment" (p. 1234) (Ramig et al., 1995). It was

reported that individuals were asked to complete self-rating forms at different intervals

during and after the study. However, the authors do not report how the self-ratings were

obtained (i.e., did the clinicians administer the self-report form, or were the clinicians

present when the individual completed the form). The bias of participants trying to give









responses that pleased the clinicians may have been a confound to the results of this

study.

Another possibility is that the variables were difficult for untrained individuals or

their families to rate in the manner intended by the investigators. Unfortunately, in the

reported comparisons of short-term (12 months) and long-term (24 months) results of the

LSVT, Ramig et al., (1996) did not report self-ratings or family ratings of perceptual

variables. If these self-ratings and family-ratings were used as outcome measures

(comparing pre-treatment and post-treatment ratings) it would have been useful to our

understanding about the effectiveness of the treatment from the subject's point of view

(the functional perspective) to have the results of the same measures at 12 months and 24

months post-treatment.

The choice of a measure of sickness-related dysfunction, such as the SIP (Bergner

et al., 1981), may also have affected their results. The SIP is a widely used measure of

dysfunction related to sickness (Pollard & Johnston, 2001). Ramig and colleagues (1996)

used two subtests of the SIP (Communication and Social Interaction) to measure

functional improvement. Individuals who received LSVT rated a significant reduction of

the impact of their illness (PD) on the communication subtest of the SIP (i.e. the affect

that PD had on their communication was reduced), while the placebo treatment group

rated no significant change in the impact of sickness on their communication immediately

after treatment (i.e., the affect that PD had on their communication was unchanged). At

12 months post-treatment the significant change reported on the SIP by the LSVT group

was not maintained, and the placebo group reported a significant increase in sickness

impact on communication (i.e., the affect that PD had on their communication had









increased, communication was more difficult) (Ramig et al., 1996). No significant

differences were found on ratings of the impact of sickness on social interactions (i.e. PD

had no affect on their social interactions), for either group, at either time.

The results seemed contradictory because many of the social-interaction items

required communication to participate in the social interactions, and one would expect

that if communication had improved there would also be improvement in the social

interaction items that pertained to communication. See Table 2-1 for examples of

communication and social interaction items from the SIP.

Table 2-1. SIP communication and social interaction items
Communication Items I do not speak clearly when I am under stress
I have difficulty speaking, for example get stuck, stutter,
stammer, slur my words
I am having trouble writing or typing
I react slowly to things that are said or done
Social Interaction Items I am cutting down on some of my usual recreation and
pastimes
I am doing fewer community activities
I am cutting down on the length of visits with friends
I am doing fewer activities with groups of people
(Bergner et al., 1981)

Review of the examples of the items on the SIP showed that the communication

items address behaviors we would classify in the ICF activity domain, while social

interaction items address behaviors we would classify in the ICF participation domain.

So although Ramig and colleagues (1995, 1996) used the SIP as an outcomes measure for

societal participation, we suggest that their results showed that the LSVT was efficacious

at the level of activity limitation, but not at the level of societal participation.

Studies by Ramig's group comprise the bulk of the treatment efficacy research for

hypokinetic dysarthria. Along with the treatment efficacy data, the authors reported that

the treatment (increasing vocal intensity) led to improved communication skills, which









led to improved quality of life for individuals with PD, though they provided no objective

evidence. We need evidence to substantiate or refute those subjective observations.

Ramig and colleagues did not report on the same measures in the study of treatment

efficacy and treatment effectiveness. They provided only subjective observations

regarding improved quality of life. With regard to treatment effectiveness they reported

that the treatment was effective based on acoustic and perceptual measures although the

SIP communication scale ratings were not maintained at either 12 months post-treatment

or 24 months post-treatment (Ramig et al., 1996).

If the goal of speech therapy, for example LSVT, is to enable a person to

communicate more effectively to participate in his life roles, measures of participation

are needed. Presently treatment effectiveness is based on measures that are applicable to

the clinical setting, but not to the social participation of individuals.

Evidence-based practice guidelines would suggest we should only administer those

treatments that improve communication effectiveness in daily life or at the participation

level. Unfortunately present measures focus on clinical efficacy versus participation

effectiveness. The results of the LSVT literature point to the need for an instrument that

can measure the communication of individuals at the level of societal participation. We

propose that the CES is such an instrument.

Measurement in Dysarthria

Kent (2000) reviewed the motor speech literature from its infancy to 2000, and

stated that while there have been instrumental and technological advances, research has

not moved far beyond the traditional studies of the classification and/or diagnosis of the

dysarthrias. Darley, Aaronson and Brown (1975) presented a systematic

description/classification of the dysarthrias by perceptual characteristics that continues to









be widely used in 2004. Their work was followed by a point-place model in which the

anatomic places where important speech activities occur are numbered and assessed by

perceptual judgments (Rosenbek & LaPoint, 1978). Beukelman and Yorkston (1977;

1979; 1981b) posited that while the work of both Darley and colleagues (1975), and

Rosenbek and LaPoint (1978) were beneficial for classification and differential diagnosis

of the dysarthrias, they did not give quantification of the person's overall speech

production. Thus they developed and standardized a method for assessing speech

intelligibility intended to give a measure of an individual's speech production of words

and connected speech (Yorkston & Beukelman, 1981a; Yorkston, Beukelman et al.,

1996; Yorkston et al., 1984).

Speech intelligibility at its simplest is defined as "the extent to which a listener

understands the speech produced by individuals with motor speech disorders." (p. 486)

(Yorkston, Beukelman, Strand, & Bell, 1999). To obtain a speech intelligibility score, a

naive judge transcribes a tape-recorded sample of words and/or sentences without benefit

of context or visual cues. The result is a percentage of intelligible words from total

words possible (Yorkston & Beukelman, 1981b; Yorkston et al., 1984). Changes in

speech intelligibility scores are traditionally reported to document treatment outcome

(Yorkston, 1996).

Speech intelligibility is a recognized measure of speech production; however,

procedures must be clearly controlled to minimize threats to the internal validity of a

study. Such threats include listener familiarization (Hustad & Cahill, 2003; Spitzer, Liss,

Caviness, & Adler, 2000), length of utterance (Spitzer et al., 2000), speaking task

(Kempler & Van Lancker, 2002), medication fluctuations (Goberman & Blomgren, 2003;









Goberman, Coelho, & Robb, 2002; Larson, Ramig, & Scherer, 1994; Mawdsley &

Gamsu, 1971; Quaglieri & Celesia, 1977), practice (Inzana, Driskell, Salas, & Johnson,

1996), and attention (Ho, lansek, & Bradshaw, 2002).

We know that sampling speech intelligibility under well-controlled circumstances

will lead to a reliable measure of an individual's speech intelligibility, and can serve as a

measure of treatment outcome per the work of Yorkston and Beukelman in their

development of the Assessment of Intelligibility of Dysarthric Speech (AIDS) (Yorkston

& Beukelman, 1981a, 1981b; Yorkston et al., 1984). However, authors report that there

is not a one-to-one relationship between a speech intelligibility score and a person's

communication success outside of the clinical treatment setting (Weismer & Laures,

2002; Yorkston, 1996).

Up to now, the lack of valid and reliable tools of functional communication (such

as communicative effectiveness) has hindered the investigation of the relationship

between speech intelligibility and functional communication. In a natural setting

communicative interactions occur in a conversational context. Communication partners

have the benefit of contextual cues, gestural cues, written cues, picture cues, and if

needed, active conversational repair, to assist with communicative effectiveness. We

concur with Ball et al., (2004) that we need valid and reliable measures to assess

successful communicative effectiveness in the participation component. We propose that

the Communicative Effectiveness Survey (CES) is such a measure.

Operationalizing Communicative Effectiveness

We define communicative effectiveness as a person's ability to actively and

efficiently get his message across successfully in his home and community settings to

fulfill life roles, using whatever means of communication possible (Hustad, 1999; Lomas









et al., 1989). Furthermore, we conceptualize communicative effectiveness as a construct

to be investigated in the participation domain of the ICF as have others (Ball et al., 2004;

Hustad, 1999; Lomas et al., 1989; Yorkston et al., 1999).

Several authors have advocated the use of a rehabilitation model such as the WHO

ICF in dysarthria research (Yorkston, 1996; Yorkston et al., 2001; Yorkston, Strand et al.,

1996). However, only two recent studies reported on the disabling nature of dysarthria in

a communicative context from the perspective of people with dysarthria. In the first, a

qualitative study of 7 participants with multiple sclerosis and dysarthria, subjects'

participation in social interactions and communication opportunities was severely

affected although they had mild communication impairments based on their speech

intelligibility scores (Yorkston et al., 2001). This is evidence that individuals focus their

concerns at the level of participation restrictions, rather than impairment or activity

limitations. The authors underscored the need for treatment goals focused on decreasing

participation restrictions as the starting point of treatment rather than the ending point

(Yorkston et al., 2001).

In a second study, Ball and colleagues (2004) reported that 25 individuals with

ALS and their caregivers rated the communicative effectiveness of the person with ALS

similarly (correlation between ratings from ALS speakers and their caregivers was r2 =

.87, p = .00). In addition, both ALS speakers and their caregivers ranked the ten items on

the CETI-M similarly with regard to level of difficulty (Wilcoxon signed ranks analysis,

no significant differences). In addition, subjects also rated their communicative

effectiveness in social situations more severely impaired although their speech

intelligibility scores were as high as 90%.









These articles indicated that individuals with dysarthria associated with multiple

sclerosis and ALS reported that they were restricted from participation in life roles due to

their communication deficits (Ball et al., 2004; Yorkston et al., 2001). We do not have

information of this sort for individuals with hypokinetic dysarthria and PD.

In summary, models of rehabilitation provide a rationale to begin developing

treatments and outcomes measures at the level of societal participation. The state of the

art in dysarthria treatment and outcomes is to use traditional, clinical measures of

perceptual judgments, acoustic, and physiologic measures to show treatment efficacy.

Those measures are often accompanied by a standard measure of connected speech called

speech intelligibility. A speech intelligibility score has worth as a measure of change at

the clinical level (termed activity component by the ICF), however it has not been shown

that the effect obtained in the clinic translates to success in an individuals' participation

in fulfilling his life roles (termed participation component by the ICF). It is imperative

that research be done at the level of participation to identify what variables are critical to

successful communication in that domain.

We propose that an instrument, such as the CES, which has demonstrated face

validity in use with individuals with aphasia and individuals with dysarthria and ALS,

may be such an instrument. However, the CES has been an instrument with unknown

psychometric properties relative to individuals with other kinds of dysarthria and

neurogenic disorders, such as those with dysarthria and Parkinson's disease.

We did a series of Rasch analyses to study the measurement properties of the CES

before using it as a measure in the current study. The next section describes the









development of the CES and the results pertinent to its usefulness as a measure of

societal participation for individuals with PD.

Development of the Communicative Effectiveness Survey (CES)

Lomas and colleagues (1989) published a report on the development, validity and

reliability of the Communication Effectiveness Index (CETI). To establish face validity

for the instrument they enlisted individuals with aphasia and their spouses to generate

items for the instrument. Next they tested the CETI using two groups of individuals; one

group (n =11) still recovering with aphasia; and one group (n =11) stable with chronic

aphasia (>12 months post-CVA). The authors reported the following internal validity

scores: internal validity was high based on Cronbach's alpha = .90; test-retest reliability

interclass correlation = .94 (95% C I = .87 to .91) and moderate inter-rater reliability

interclass correlation =.73 (95% CI = .62 to .81). In addition, construct validity was

determined using Spearman rank correlation coefficients between the CETI and three

measures of aphasia. Their results indicated that the CETI showed a change in

communicative effectiveness pre- and post-treatment for individuals with aphasia. We

performed a post hoc analysis of power and effect size (t = 3.68, df = 20, d= 1.57; 95%

CI =.6 < 1.57 > 2.5) to determine the statistical rigor of the instrument. Power of .94 for

effect size of .8 indicated adequate statistical power of the results, although the sample

size was small.

Yorkston et al., (1999) modified Lomas and colleagues' (1989) CETI from 14 to 8

items to be used by individuals with dysarthria. Their first modification of the CETI was

to delete the items that pertained to language that did not pertain to individuals with

motor speech disorders. Next, they involved individuals with dysarthria in the item

selection process. The apparent usefulness and face validity of the instrument without the









necessary statistical analyses was the impetus for two stages of Rasch analysis to

establish the measurement characteristics of the CES.

Rasch analysis, a single-parameter model of item response theory, was used to

determine the measurement properties of the CES using computer software called

WINSTEPS Ministep (Linacre, 1991). Rasch analysis is designed to provide an

investigator with information about the important constructs of measurement, such as the

initial content validity of the instrument, reliability of responses of the participants,

unidimensionality of items, along with how well the items fit together to define a

theoretical construct (such as communicative effectiveness). In addition, Rasch analysis

transforms ordinal data (counts) into interval data, which results in a log-linear measure

of item difficulty, and person ability calibrated on a single scale (Bond & Fox, 2001;

Rasch, 1980; Wright, 1997; Wright & Linacre, 1989). Finally, Rasch analysis generates

information on ceiling and floor effects, indicators of whether or not the instrument had a

sufficient level of difficulty to accommodate the range of the population the instrument

was designed to measure (Velozo & Peterson, 2001).

In the first stage of analysis to determine the measurement characteristics of the

CES a speech-language pathologist blind to the purpose and hypotheses of the study

administered the CES, an eight-item, seven-point interval scale, to 95 individuals with

movement disorders (diagnosed by a movement disorders specialist from the University

of Florida Movement Disorders Center). The individuals with movement disorders had

speech competencies ranging from normal to severe dysarthria. Table 2-2 gives the

demographic data for the group.









Table 2-2. Demographics of participants in two Rasch analyses of CES ratings
All Movement Disorders) PD only
Mean age 62.6 (range 18-89) 66.5 (range 49-87)
Gender distribution 58% Male, 42% Female 64% Male, 36% Female
Diagnosis PD 70%, Dystonia 10%, Other 20% PD 100%
Speech competence Normal to severe dysarthria Normal to severe dysarthria

We hypothesized that the CES would show measurement properties sufficient for

use as an instrument to measure the communicative effectiveness of people with

movement disorders, as it had with individuals with aphasia. Rasch analysis was used to

determine the measurement properties of the instrument by converting the ordinal data to

interval data using WINSTEPS, a computer software program (Linacre, 1991). See

Table 2-3 for a summary of the Rasch analysis.

The results of the first analysis showed that the CES had sufficient measurement

properties, along with initial construct validity and response reliability, to continue

investigating its usefulness as a measure of communicative effectiveness for individuals

with movement disorders and a wide range of speech competence. The only exception to

our findings was that individuals used a 5-point rating scale more effectively than a 7-

point scale. The 5-point scale was used for all further analyses.

The purpose of the second stage of analysis was to analyze the measurement

properties of the CES for individuals with PD (70% of the original sample). See Table 2-

2 for the group demographics. We hypothesized that the CES would show similar

measurement properties for individuals with PD, as it showed for individuals with

various movement disorders. The results showed that the second analysis were indeed

similar to the first, with the exception that individuals with PD used a 4-point rating scale

more effectively than the 7-point scale. See Table 2-3 for a summary of the second

Rasch analysis.










Table 2-3. Results of two Rasch analyses of CES ratings
All movement disorders
Sample size >50a N=95

Person-to-Measure Correlation* .90

Unidimensionality
(MnSq=l Ideal)
In-fit Statistics Persons MnSq = .99
In-fit Statistics Items MnSq = .98

Levels of Difficultybc 3.13 statistically significant


Actual Item hierarchy agreement
with theoretical item hierarchy


it


levels of ability

7/8 (+1 item)
Hierarchy of items agreed with
the hierarchy proposed a prior


PD only
N=66

.90



MnSq = .99
MnSq = .98

3.05 statistically significant
levels of ability

6/8 (+1 item)
Hierarchy of items agreed
with the hierarchy proposed a
prior


Redundancy of items 1 1
Talks when upset rated the Talks when upset rated the
same difficulty as talks in a same difficulty as talks in a
noisy environment noisy environment

Reliable use of rating units Used a 5 point rating scale Used a 4 point rating scale
more reliably than a 7 point more reliably than a 7-point
scale scale
a(Linacre, 1994), b(Wright & Stone, 1979), c(Velozo & Peterson, 2001) Comparable to
Cronbach's alpha

To determine that the 4-point rating scale was more effective than the 5-point rating

scale, we used a set of 8 guidelines presented by Linacre (2002). We applied the

guidelines to the summary of measured steps to determine the effectiveness of the 4-point

rating scale. See Table 2-4. Our results met 7 of 8 of the criteria that Linacre established

as essential for rating scale use. See Table 2-5 for those results. The exception was that

there was not a regular distribution in our category count.

Based on both populations analyzed, the CES showed sufficient measurement

properties, along with initial face validity, content validity, and reliability of response

(based on person-to-measure correlation, similar to Cronbach's alpha), to continue










investigating its usefulness as a tool to quantify communicative effectiveness of

individuals with PD and a wide range of speech competence.

Table 2-4. Summary of measured steps 4-point rating scale PD sample
OUTFIT Step Coherence Coherence Zone Zone
Category Average Expected MnSq Calibration M- C% C-M% From To
Count Measure Measure *4 *5 *6 *6 *7, 8 *7, 8
*1, 2 *3 *3 (<2.0) (>1 logit) (>40%) (>40%) (>1.4) (<5.0)
1 28 -4.02 -4.09 1.03 72% 46% -Inf. -4.50
2 179 -1.07 -1.04 1.03 -4.48 73% 75% -4.50 .49
3 166 1.95 1.94 .89 .57 64% 72% .49 4.00
4 87 4.47 4.45 1.04 3.91 78% 64% 4.00 +Inf.
Note: *Number identifies pertinent guideline in Table 2-5

Table 2-5. Summary of CES 4-point Rating Scale Characteristics of PD Sample


Guideline
Pre. Scale oriented with
latent variable

1 At least 10 observations
of each category

2 Regular Observation
distribution

3 Average measures
advance monotonically
with category

4 OUTFIT mean-squares
less than 2.0

5 Step calibrations


6 Ratings imply
measures, and measures
imply ratings

7 Step difficulties
advance by at least 1.4
logits


Measure
Stability
(E)
MET


(E)
MET


Measure
Accuracy
(E)
MET


Description of
this sample
(E)
MET


(H)
MET


Inference for
next sample
(E)
MET


(H)
MET


(H)
UNMET


(H)
MET


(H)
MET


(H)
UNMET


(E)
MET


(E)
MET


(E)
MET


(H)
MET


(H)
MET


(E)
MET


(H)
MET

(H)
MET

(H)
MET


(H)
MET


8 Step difficulties (H)
advance by less than 5.0 MET
logits
(Linacre, 2002) Note: (E) = Essential; (H)
guideline


Helpful, Blank areas


not applicable to the















CHAPTER 3
DESIGN AND METHODS

This study was designed to test the face validity, content validity, and construct

validity of the CES as a measure of societal participation for individuals with dysarthria

and PD. To accomplish that goal we designed a prospective study with mixed statistical

design.

To test Hypothesis 1 a case-control, one-directional paired samples t-test, a = .05

was used to investigate whether the CES ratings of healthy community-dwelling speakers

were significantly higher than the CES ratings of speakers with dysarthria and PD. To

test Hypothesis 2 a one-directional dependent samples (familial blocking) t-test, a = .05

design was used to investigate whether the CES ratings of 25 participants with dysarthria

and PD was significantly higher than the CES ratings given to them by their significant

others (SO). To test hypothesis 3, a multiple regression analysis was used to determine

the relationships between the CES and the two measures of speech intelligibility of

speakers with dysarthria and PD.

The study required recruitment of three groups of subjects. The Parkinson's

disease group designated PD, included individuals diagnosed with PD who also had

dysarthria. The non-PD group designated NO, included individuals with no existing

neurologic injury or disease and no dysarthria, who were matched on gender and mean-

age to the PD group. The significant others group designated SO, included the significant









others of the individuals in the PD group. Significant other was defined as the person the

PD individual identified as a primary communication partner.

Participants

We enrolled 25 participants in each of three groups for a total of 75 subjects, based

on an a priori power analysis for a one-tailed t-test using a = .05, and power = .8. (Faul &

Erdfelder, 1996). All participants were community-dwelling volunteers recruited from

the University of Florida, Department of Neurology Movement Disorders Clinic, PD

support groups, or elder community centers.

For all subjects, exclusion criteria included a reported history of neurologic,

psychiatric, or language deficit, Mini-mental State Exam (MMSE) of less than 23

(Folstein, Folstein, & McHugh, 1975), Apathy Scale less than 14 (Starkstein et al., 1992),

non-reader, non-English speaker, unable to follow directions due to hearing loss, less

than 18 or more than 90 years of age. Individuals with PD also had to have a Hoehn

&Yahr Staging for Parkinson's disease (H&Y) stage between 1 and 4. The H&Y is a

staging instrument that is used in PD research which assesses physical ability. An

example of a stage 1 item is symptoms are inconvenient but not disabling. An example

of a stage 5 item is requires constant nursing care (Hoehn & Yahr, 1967). See Appendix

B for all of the items on the scale. The participants also had to have a rating between 1

and 4 on the Dysarthria Rating of Severity (DRS). The DRS is a scale that rates the

severity of dysarthria from stage 1, no detectable speech disorder, to stage 5, no

functional speech (Yorkston et al., 1999). The healthy individuals and significant others

were excluded if they had a history of a speech impairment. Table 3-1 gives the

complete listing of inclusion and exclusion criteria for each group.









Table 3-1. Inclusion/exclusion criteria for study participants by group


Group Inclusion Criteria
PD Diagnosis of PD > 6 months
n=25 Pharmacologically stable
Can ID on/off movement cycles
Hoehn & Yahr Staged 1-4 a
MMSE > 23b
Apathy Scale <14c
Dysarthria Functional Limitation
Stage 1-4d
Functional sentence reading skill
Native English speaker
Hearing acuity sufficient to follow
directions
Visual acuity sufficient to read large
print
Community dwelling
18 to 90 years old

NO No diagnosed PD
n=25 MMSE >23b
Apathy Scale < 14
No Dysarthria
Functional sentence reading skill
Native English Speaker
Hearing acuity sufficient to follow
directions
Visual acuity sufficient to read large
print
Community dwelling
18 to 90 years old

SO No history of neurologic, psychiatric,
n=25 speech, or language deficits
MMSE >23b
Apathy Scale < 14'
Functional sentence reading skill
Native English Speaker
Hearing acuity sufficient to follow
directions
Visual acuity sufficient to read large
print
Community dwelling
18 to 90 years old


Exclusion criteria
History of other neurologic,
psychiatric, or language deficits.
Hoehn & Yahr staged >4a
MMSE < 23b
Apathy Scale >14c
No dysarthria4
Cannot read
Non-English speaker
Severely HOH
Visually impaired
Non-community dwelling
No SO, or SO who is unwilling to
participate
<18 and >90 years old



History of PD, other neurologic,
psychiatric, speech, or language
deficits
MMSE <23b
Apathy Scale > 14'
Cannot read
Non-English speaker
Severely HOH
Visually impaired
Non-community dwelling
<18 and >90 years old


History of neurologic, psychiatric,
speech, or language deficits
MMSE <23b
Apathy Scale >14c
Cannot read
Non-English speaker
Severely HOH
Visually impaired
Non-community dwelling
<18 and >90 years old


a(Hoehn & Yahr, 1967) b(Folstein et al., 1975), c(Starkstein et al., 1992), d(Yorkston et
al., 1999)









Recruitment

Parkinson's Disease Group

The Research Coordinator at the UF Movement Disorders Center (UF MDC),

Gainesville, FL, who was trained to identify and recruit subjects for clinical trials in PD

according to standards established by the NIH, approved a query of the UF MDC

Database to identify individuals who met the study eligibility criteria to participate in our

study (based on IRB-approved Informed Consent #416-2002). The Principal Investigator

(PI) made the initial contact with potential participants. If the individual indicated a

willingness to participate, the PI conducted a telephone screening to ensure that the

individual met the inclusion/exclusion criteria. If the individual met the

inclusion/exclusion criteria, an appointment was scheduled to obtain informed consent,

perform the screening, and if appropriate, perform the study.

Non-PD Group

Participants were recruited by the PI through responses to an IRB-approved printed

flyer distributed to communities of older individuals. When a participant responded, the

PI conducted a telephone screening to determine whether or not the participant met the

inclusion/exclusion criteria for the study. If the individual met the inclusion/exclusion

criteria an appointment was scheduled to obtain informed consent, perform the screening,

and if appropriate, data collection.

Significant Other Group

One of the eligibility criteria for the subjects with PD was that they had a

significant other who was also willing to participate in the study. If the PD subject had a

significant other willing to participate, a telephone screening was conducted with the









significant other at the same time the PD participant was contacted. Appointments for

the PD participant and SO participant were scheduled for the same time.

Sample Demographic Data

Using the inclusion/exclusion criteria listed in Table 3-1, and the described

recruitment procedures, the 75 individuals were recruited into the study. The

demographic data resulted in the following group compositions. The PD group was

comprised of 25 individuals with dysarthria and PD, 14 males and 11 females. Their

average age was 70.3 years (range 60-84 years, SD = 6.29 years). See Table 3-2 for

individual demographics.

The non-PD group was made up of 25 individuals, 14 males and 11 females. Their

average age was 70.4 years (range 44-79 years, SD = 8.69 years). See Table 3-2 for 1

demographics of the NO group.

The SO Group was made up of 25 individuals, 10 males and 15 females. SOs were

the spouses of individuals with Parkinson's disease in 21 of the 25 cases (84%). The

other four participants included one sibling, two children, and one primary

communication partner. See Table 3-2 for the demographics for the SO group.

Informed Consent

The PI obtained informed consent for all seventy-five subjects, using the

procedures established and approved in IRB#276-2004. The PI had received all of the

required training in the proper procedures for obtaining informed consent and protecting

the individual's health care information for both HIPAA and NIH compliance.










Screening Measures

The inclusion and exclusion criteria established to control the threats to the external

validity of the study required that participants complete a short screening battery prior to

completion of the study. These measures and the possible threats to validity included:

*Mini Mental State Examination (MMSE). Cognitive deficits including dementia are
typically associated with the overall severity of PD, and may be part of other diseases
and aging as well. A short cognitive screening was chosen to control for the
unreliable responses that might result from someone with diminished cognitive skills
in all three groups. The MMSE is widely recognized as a quick screening tool that
gives a gross measure of a person's cognitive skills (Folstein et al., 1975). We
adhered to the MMSE's established cutoff score of < 23 as an indicator of impaired
cognition. If the participant obtained a score of 23 or less he/she was excused from
further participation in the study.


Table 3-2. Demographic data for all groups
Group PD Group
Participant Age Gender
01 76 M
02 69 F
03 69 M
04 75 M
05 72 M
06 68 F
07 69 M
08 66 F
09 63 M
10 62 M
11 77 F
12 69 F
13 60 M
14 68 M
15 73 M
16 74 M
17 68 F
18 65 F
19 69 M
20 60 F
21 84 F
22 71 F
23 84 M
24 69 M
25 77 F


NO Group
Gender
F
M
F
M
F
M
F
F
F
M
F
F
M
F
M
F
M
M
M
M
M
M
M
M
F


SO Group
Gender
F
F
M
F
F
M
F
M
F
F
M
M
F
F
F
F
M
F
F
M
M
M
F
F
M









* Apathy Scale. An apathy rating was included because there is evidence that apathy
may occur in individuals diagnosed with PD (with and without the presence of
depression) (Marin, 1990; Starkstein et al., 1992). Apathy may preclude successful
participation in life roles (the domain we intended to investigate in our study).
Depression, common to individuals with PD, is routinely recognized and treated in
the UF Movement Disorders Center; therefore a depression screening was not
included in this screening battery. Apathy is defined as a lack of motivation that is
not attributable to diminished level of consciousness, cognitive impairment, or
emotional distress (Marin, 1990; Starkstein et al., 1992). The scale used a cutoff
score of>14. See Appendix C for the Apathy Scale.

* Hoehn and Yahr Staging for Parkinson's disease (H&Y). We determined that the
study would include individuals between H&Y stage 1 and stage 4 to meet the criteria
that they be community-dwelling (Hoehn & Yahr, 1967). The H&Y is a five-stage
scale designed to rate the disability associated with Parkinson's disease. The stages
range from one to five, with one being the mildest (mild unilateral symptoms,
inconvenient but not disabling, friends notice changes in posture, locomotion and
facial expression) and five being the most severely disabled (complete invalidism,
constant nursing care required, cannot walk or stand). See Appendix B for the
complete scale.

* Dysarthria Rating of Severity (DRS). We determined that the study would include
individuals who had stage 1 dysarthria severity (speech problems and increased effort
to speak noticeable to the individual), through stage 4 dysarthria severity (severe
dysarthria, minimal speech intelligibility and frequent communication breakdowns).
The DRS is a 5-stage model that clinicians can use to rank the reported and observed
severity of dysarthria of individuals with PD (Yorkston et al., 1999). If an individual
with PD was observed (and self-reported) to have normal speech, he was excused
from participating in the study. Table 3-3 describes the DRS in detail.

Table 3-3. Dysarthria Rating of Severity
Stage Description
Stage 1: No detectable Individuals who place high demand on communication
speech disorder will report changes in speech performance
Complaints include sense of increased effort to speak

Stage 2: Obvious speech Changes in voice: reduced loudness, reduced pitch
disorder with intelligible variability and unsteady voice
speech Complaints include difficulty speaking in specific
communication contexts, such as a noisy environment

Stage 3: Reduction in Frequently asked to repeat
speech intelligibility Imprecise oral articulation
Limited range of oral movements
Rushes of speech with abnormal placement of pauses









Table 3-3. Continued
Stage Description

Stage 4: Natural speech Natural speech alone is no longer functional
supplemented by Frequent communication breakdowns
augmentative techniques Severe dysarthria with minimal speech intelligibility

Stage 5: No functional Unable to speak at all
speech Requires augmentative and assistive communication
devices
(Yorkston et al., 1999)

Dependent Measures

* Communication Effectiveness Survey (CES). The CES is the 8-item, 4-point equal
interval scale developed in our laboratory. The CES showed initial measurement
properties that suggested it would be a useful measure of change in communicative
effectiveness of individuals with the dysarthria of PD. See Appendix A for the CES.

* Sentence Intelligibility (SIT) of the Assessment of Intelligibility in Dysarthria
Speech (AIDS). The AIDS has standardized instructions for administration, along
with established validity, and reliability (Yorkston & Beukelman, 1981a, 1981b;
Yorkston et al., 1984). It is a measure of connected speech intelligibility elicited by
asking a participant to read a set of sentences aloud.

* Spontaneous Speech Intelligibility (SSI). Kempler & Van Lanker (2002) showed
that a speech intelligibility measure based on a reading sample resulted in
significantly higher speech intelligibility scores than scores derived from a
spontaneous speech samples of individuals with the dysarthria of PD, although many
other investigators have shown that spontaneous speech intelligibility tasks result in
significantly higher scores because listeners can derive more from the context of
connected speech (Hustad & Beukelman, 2002; Hustad & Beukelman, 2001; Hustad
& Cahill, 2003) Because of these conflicting results and because intelligible,
spontaneous speech is the ideal outcome of speech therapy we added a measure of
spontaneous speech intelligibility to our study. Each individual in the PD group was
asked to talk for one minute about the place he/she was born after completing the SIT
task.

Procedures

Screening Procedures

Once a telephone screening confirmed the individual's eligibility to participate in

the study, an appointment was set up at the sight of the participant's choice: home, VA

Speech Pathology Clinic, or Shands Speech and Hearing Clinic. Informed consent was









obtained prior to the start of every study. It was obtained in the privacy of the

individual's home or in a private treatment room where the participant could ask and

answer questions confidentially.

After individuals signed the informed consent, the MMSE and Apathy Scale were

administered. In addition the individuals diagnosed with PD received the H&Y staging

and the DRS rating. The PI administered and scored all screening instruments.

Individuals who did not pass the screening were thanked for their time and excused.

Individuals who passed the screening continued on to participate in the study.

Dependent Measure Collection Procedures

The experiment took place in a quiet room of the participant's choosing. Seventy-

one participants chose to have the study conducted in their home. The remainder

completed the study in the clinic

For all participants with PD, the experiment began 30 minutes after they had taken

their medications or were in an "on" medication cycle, and ended 30 minutes before they

were due to take their medication again or were in an "off' medication condition. We

chose to assess participants in the PD group in an "on" medication cycle to obtain

optimal speech intelligibility scores to compare with the communicative effectiveness

ratings, recognizing that when individuals are in an "off' medication cycle speech

intelligibility would be decreased. For the participants without PD the study was

scheduled at their convenience.

All participants completed the CES independently after receiving verbal and

written instructions. After the CES was completed every subject was asked to give

his/her impressions about the instrument. The PI noted those impressions. NO and SO

groups were excused after the debriefing.









Following completion of the CES, the PD group completed the SIT by reading 10

sentences that were 14 words in length. Twenty-five unique sets of 10 fourteen-word

sentences were developed from the 100 14-word sentences presented in the AIDS to

control for listener familiarity (Yorkston et al., 1984). Research Randomizer, a web-

based random number generator was used to obtain the 25 unique sets of numbers

(Urbaniak & Plous, 1997. http://222randomizer.org).

After the sentence intelligibility task was completed each PD participant generated

a one-minute spontaneous speech intelligibility (SSI) sample based on the directions, "I

would like you to talk to me for one minute about the place you were born and grew up.

Please keep talking until I tell you to stop."

Scoring Dependent Measures Procedures

Three native English speakers who had no hearing loss (based on self-report)

transcribed the Sentence Intelligibility (SIT) and Spontaneous Speech Intelligibility (SSI)

samples. The transcribers were given both printed and verbal instructions from the AIDS

for transcribing the samples. Transcribers had no familiarity with or access to the

sentence stimuli used. As stated above, sentences were randomized for each participant

to decrease the transcribers' familiarity with the sentence intelligibility samples. The PI

scored all SIT and SSI transcripts for intelligible and unintelligible words, per AIDS

instructions (Yorkston & Beukelman, 1981a). The SIT and SSI score used for statistical

analysis was an average of the scores of the three judges.

Data Collection Methodology

The following methods were used to collect, store and transcribe the dependent

measures:









* Published AIDS manual instructions for test administration and sample collection
were followed (Yorkston et al., 1984).

* A headset noise-resistant microphone (TS031), positioned 2 cm from the speaker's
mouth captured the speech samples.

* Microsoft MS Sound Recorder system contained in the Dell Inspiron 8600 Laptop
Computer was used to record the speech in PM format, at 48.000 kHz, 16 Bit, stereo
at 187 kb/sec.

* Background noise was < 30 dB-A measured by a sound level meter throughout
recording. The sound level meter was calibrated prior to each recording.

Each subject was seated in an upright position, and the microphone was positioned.

Each subject received the following instructions in an effort to minimize reading errors:

"Please read the sentences to yourself one time. When you are ready, I will ask you to

say the number behind the period and then read the sentences in your usual speaking

voice. Please pause between each sentence." Individual samples were saved to the

computer hard drive, using a patient identification number and date. Recording and

transcription parameters were the same as for the SIT. The speech samples of the 25 PD

individuals were recorded onto three high fidelity CD's for transcription purposes. The

transcribers used "sound" computer software and free field listening in a quiet setting

with the listening volume set at a comfortable level to transcribe the samples. Their

handwritten responses were returned to the PI for scoring. To ensure that the recordings

were of sufficient quality, the three judges were asked to rate the quality of the speech

recording on the CD's they received as excellent, good, fair, and poor. All three rated the

recording quality as good.

Inter-rater Reliability

Three judges transcribed the sentences and the spontaneous speech samples of the

25 individuals with dysarthria and PD. To assess the reliability of three judges' ratings,









Cronbach's a was used to compare the variation of ratings for a particular judge in

relation to the co-variation in ratings for all three judges. For intelligible words on the

SIT task, the corrected judge to total correlations were .94, .96, and .86 for the three

judges respectively yielding a standardized reliability of a = .97. For unintelligible words

on the SIT task, the corrected judge to total correlations were .84, .94, .82 yielding a

standardized reliability of a = .95. For intelligible words on the SSI task, the corrected

judge to total correlations were .98, .96, and .91 for the three judges respectively yielding

a standardized reliability of .98. Finally, for unintelligible words on the SSI task, the

corrected judge to total correlations were .69, .61, and .55 for the three judges

respectively yielding a standardized reliability of .83. Table 3-6 shows the results of the

test for inter-rater reliability. To assess accuracy of scoring, a third party blinded to the

study's purpose transcribed five randomly selected SIT and SSI samples. Accuracy of

the transcriptions was 96% agreement between the PI and the third party.

Table 3-4. Inter-rater reliability
Standardized
Task Judge 1 Judge 2 Judge 3 Reliability
SIT intelligible words .94 .96 .86 a =.97
SIT unintelligible words .84 .94 .82 a .95
SSI intelligible words .98 .96 .91 a .98
SSI unintelligible words .69 .61. .55 a -.83
Note: SIT=Sentence Intelligibility; SSI=Spontaneous Speech Intelligibility

Data Analysis Plan

SPSS was used to produce descriptive statistics and the following inferential

statistical tests:

*Hypothesis 1: The CES ratings for 25 participants with dysarthria and PD will be
significantly less than CES ratings of 25 mean gender- and age-matched healthy
participants. This hypothesis was tested using a one-tailed dependent samples t-test
with the Type 1 error rate set at a =.05.









* Hypothesis 2: The CES ratings of 25 participants with dysarthria and PD will be
significantly larger when compared to the CES ratings given to them by their
significant others (SO). Due to the dyadic relationship between the speakers and
significant others, this hypothesis was tested using a one-tailed dependent samples t-
test with the Type 1 error rate set at a = .05.

* Hypothesis 3: When predicting sentence intelligibility score (SIT) from
communicative effectiveness survey score (CES), spontaneous speech intelligibility
score (SSI) accounts for a significant amount of the variation in SIT above and
beyond the variation due to CES for individuals with dysarthria and PD. Likewise,
when predicting SSI from CES, SIT accounts for a significant amount of the variation
in SSI above and beyond the variation due to CES for individuals with dysarthria and
PD. This hypothesis was tested using one-directional multiple regression and
comparison of squared semi-partial correlation coefficients. Type 1 error rate set at a
= .05. The dependent measure was the CES, regressed on SIT and SSI scores.













CHAPTER 4
STUDY RESULTS

* Hypothesis 1. The CES ratings for 25 participants with dysarthria and PD will be
significantly less than CES ratings of 25 gender- and mean-age-matched healthy
participants.

Results of Testing Hypothesis 1

This hypothesis was tested using a one-tailed dependent samples t-test with the

Type 1 error rate set at a =.05. A paired-samples t-test was used to analyze the mean

difference between the PD group and the NO group. The paired-samples test was

justified by prior matching of participants between conditions on gender and mean age.

The per comparison error rate was controlled at a = .05. The correlation between groups

was .083 and was not significantly different from zero (p = .69). Healthy individuals

(NO group) had significantly higher CES ratings (M = 28.68, SD = 2.82) than participants

with dysarthria and PD (M= 23.24, SD = 4.82) with a mean difference of 5.44 between

groups, t (24) = -4.70, p < .001. The 95% confidence interval around the mean difference

between groups was 3.05 and 7.83. These data give a measure of construct validity to the

CES. See Figure 4-1.

* Hypothesis 2: The CES ratings of 25 participants with dysarthria and PD will be
significantly higher when compared to the CES ratings given to them by their
significant others (SO).

Results of Testing Hypothesis 2

This hypothesis was tested using a one-tailed dependent samples t-test with the

Type 1 error rate set at a = .05. This analysis compared the mean difference on the CES

ratings for individuals with Parkinson's disease (PD group) and the significant others'







40


(SO Group) CES ratings of the individuals with PD. Although there was a moderate

correlation between group (r = .69), the relationship was not significantly different from

zero. The results of the t-test indicate that the PD group participants had significantly

higher CES ratings (M= 23.24, SD = 4.82) than their CES ratings given by the significant

others (SO group) (M= 21.64, SD = 5.80) with a mean difference of 1.60, t (24)= 1.864,

p =.037. See Figure 4-2.


r30








12
li
Io



4

0
Mean CES Mean CES
Rating-PD Group Rating-NO Group

Figure 4-1. Comparison of means for PD and NO groups

*- -1 *-----------------------


Mean CES Mean CES
Rating-PD Group Rating-SO Group
Figure 4-2. Comparison of means for PD and SO groups









*Hypothesis 3: When predicting sentence intelligibility score (SIT) from
communicative effectiveness survey score (CES), spontaneous speech intelligibility
score (SSI) accounts for a significant amount of the variation in SIT above and
beyond the variation due to CES for individuals with dysarthria and PD. Likewise,
when predicting SSI from CES, SIT accounts for a significant amount of the variation
in SSI above and beyond the variation due to CES for individuals with dysarthria and
PD. This hypothesis was tested using one-directional multiple regression and
comparison of squared semi-partial correlation coefficients to determine the
relationships among the three measures. The dependent measure was the CES, the
independent measures were the SIT and SSI scores The multiple regression formula
for this hypotheses was: CES = a + bsitXsit + bssiXssi..

Results of Testing Hypothesis 3

Standard multiple linear regression analysis was performed using SPSS

REGRESSION to determine the regression coefficients for CES, SIT and SSI.

Assumptions were tested by examining normal probability plots of residuals and scatter

plots of residuals versus predicted residuals. No violations of normality, or linearity were

found. SPSS REGRESSION gives the regression coefficients 0, p-intervals, R2, F andp-

values for the regression analysis. The full model CES = 2.94 + 47.27(SIT) 25.46 (SSI)

was not significant, F (2, 22) = 1.97, p = .163. R2 = .152, indicated that SIT and SSI

accounted for only 15% of the variability in the observations. Table 4-1 gives the

statistical data used in the regression model.

Table 4-1. Summary of variables SIT and SSI contributing to change in CES
Unstandardized Standardized
Coefficients Coefficients 95% Confidence Interval
Model P Std. Er. 3 t Sig Lower Bd. Upper Bd.
(Constant) 2.94 16.01 .18 .86 -30.24 36.14
Avg. % SIT 47.27 27.16 .60 1.74 .10 9.06 103.50
Avg. % SSI -25.45 28.80 -.30 .88 .39 -85.18 34.27
Note: CES=Communicative Effectiveness Survey; SIT=Sentence Intelligibility; SSI=
Spontaneous Speech Intelligibility

The partial correlation coefficients (0) indicated that approximately 15% of the

variance of SIT and SSI (11.6% and 3% respectively) was accounted for in the

observations, leaving 85% of the variance unaccounted for. This result could be









attributed to random variability or it could indicate that we did not account for other

variables that affected CES ratings. This result led to the following post hoc analyses.

Post Hoc Analyses

In attempting to discover what other variables contributed to changes in CES

ratings, a review of mean CES ratings by H&Y indicated that the PD group rated

decreases in CES in relation to increased physical disability and increased dysarthria

severity. We felt that post hoc analysis of these severity data was warranted to

understand how individuals with dysarthria and PD rated their communicative

effectiveness in relation to the severity of their physical deficits and the severity of their

dysarthria. See Figure 4-3.


30

25

20





5

0
1.00 2.00 3.00 4.00
Hoehn & Yahr Stages of PD
Figure 4-3. CES ratings for H&Y stages of PD

We began our post hoc analysis by reviewing our sample characteristics. First, we

noted that our sample was skewed toward individuals with less severe physical disability,

based on the H&Y ratings (M = 2.08, SD= .996, range 1-4) (where 1 is mild and 5 is

total invalidism). Although it was skewed, the sample met the necessary assumptions for

statistical tests used including linearity, and normality (test of kurtosis was not









significant). The same held true for the DRS (M = 2.16, SD = .62, range 1-3) (where 1 is

very mild and 5 is no usable speech). See Figure 4-4. Finally, both sentence

intelligibility (SIT) and spontaneous speech intelligibility (SSI) scores neared the ceiling

of 100% intelligible (SIT [M= 94.2%, SD= 6.10%], and SSI [M= 95.4%, SD= 5.78%]).

See Table 4-3 for descriptive statistics.



30.


-2
Z 20



10



1.00 2.00 3.00
Dysarthria Rating of Severity
Figure 4-4. CES ratings for ratings of dysarthria severity

Standard multiple linear regression analysis was performed between CES

(dependent variable) and the SIT, SSI, H&Y, and DRS using SPSS REGRESSION

model CES = a + bsitXsit + bssiXssi. + bh&yXh&y + bdrsXdrs. Assumptions were tested by

examining normal probability plots of residuals and scatter plots of residuals versus

predicted residuals. No violations of normality, or linearity were found. Table 4-4- gives

the complete statistical data for the multiple regression analysis of CES, SIT, SSI, H&Y

and DRS.









Table 4-2. Descriptive statistics for CES, SIT, SSI, H&Y & DRS
Measure Mean Standard Deviation Number
CES Total 23.24 4.82 25
Average %SIT .94 .06 25
Average %SSI .95 .06 25
H&Y 2.08 .99 25
DRS 2.16 .62 25
Note: CES=Communicative Effectiveness Survey; SIT=Sentence
Intelligibility; SSI= Spontaneous Speech Intelligibility; H&Y=Hoehn
& Yahr Staging of Parkinson's disease (Hoehn & Yahr, 1967);
DRS=Dysarthria Rating of Severity (Yorkston et al., 1999)

Table 4-3. Summary of variables SIT, SSI, H&Y, and DRS contributing to change in CES
Unstandardized Standardized
Coefficients Coefficients 95% Confidence Interval
Model B Std. Er. Beta t Sig Lower Bd. Upper Bd.
(Constant) 48.98 15.66 3.13 .005 16.32 81.64
Avg. % SIT 21.04 20.92 .27 1.01 .327 -22.60 64.68
Avg. % SSI -37.02 27.71 -.44 -1.08 .104 -82.31 8.27
H&Y 2.72 1.24 -.56 -2.19 .040* 5.30 .13
DRS 2.13 1.81 -.28 -1.17 .255 5.91 1.66
Note: CES=Communicative Effectiveness Survey; SIT=Sentence Intelligibility; SSI=
Spontaneous Speech Intelligibility; H&Y=Hoehn & Yahr Staging of Parkinson's
disease (Hoehn & Yahr, 1967); DRS=Dysarthria Rating of Severity (Yorkston et al.,
1999

The full model was significant (F[4, 20] = 6.88,p = .001, R2 = .588), and 59% of

the variability in the CES was accounted for by the revised model. Analysis of the

standardized coefficients (beta) showed that the H&Y variable was significant (t = -2.19,

p = .040) in this model and accounted for the largest portion of variance of all the

variables. It was followed by the SSI, which approached significance (t = -1.08, p =.10).

These results suggested that as individuals declined in physical ability, we could predict a

decline in their communicative effectiveness (participation component measure) as well,

although their mean speech intelligibility scores (impairment component measure)

remained high. This result adds evidence to the content validity of the CES as a measure

of communication in societal participation.









Participant Feedback

After subjects had completed the CES they were asked to respond to three

questions, about the survey, and were free to answer or not. The questions were: 1)

Overall, what is your impression of the CES? 2) Based on your experience, as a person

with dysarthria and PD or the significant other of a person with dysarthria and PD are

there other social speaking situations we should add to this survey? 3) Is there any other

information about your communicative effectiveness that you think other healthcare

professionals should know? The PI noted their comments on a debriefing form as they

were talking. The PI asked for clarification if a comment was unclear. Of the 50 who

participated in the study 25 gave comments during the debriefing other than "I thought it

was fine" or "I wouldn't add anything." See Appendix C for a listing of the comments

given by both individuals with dysarthria and PD, and their significant others.

In general, participants appreciate the brevity of the survey. They also reported that

the survey tapped most of their daily communication situations. Suggestions for added

items included: speaking when fatigued, speaking when off meds versus on meds,

speaking when there is increased noise in the home, speaking when anxious, rather than

angry or upset, speaking when engaged in a physical activity, speaking outdoors,

speaking to a group, speaking when on the spot, and speaking for a long period of time.

In summary, we were able to accept Hypothesis 1 and Hypothesis 2. The results of

Hypothesis 3 testing were more complex and required post hoc analysis to discern the

salient predictors of CES. We were unable to accept our hypothesis that SIT and SSI

would be significant predictors of change in CES ratings. Post hoc analysis revealed that

the only significant predictor of change in CES ratings (based on our model SIT,SSI,

H&Y, and DRS) was the H&Y, a measure of physical disability, followed by






46


spontaneous speech intelligibility. The contributions of these results to the field of

dysarthria assessment, treatment, and research will be discussed in the next section.














CHAPTER 5
DISCUSSION

We began this study by proposing that a measure of communication in the ICF

societal participation component was needed for the field of motor speech disorders

because although researchers often assume that the treatment effects obtained in the clinic

translate to success in an individual's societal participation to fulfill life roles, they have

not often shown it. It is impossible to investigate the phenomenon of communication in

societal participation at this time because a psychometrically sound measure of

communication in the societal participation component does not yet. We proposed that

the Communicative Effectiveness Survey, a survey which measures communicative

effectiveness in 8 everyday speaking situations, may serve as a measure of

communication in societal participation, based on the literature, and our early

foundational work in analyzing the measurement properties of the CES

The purpose of this study was to investigate three types of validity, face validity,

content validity, and construct validity, of the CES for individuals with dysarthria and

PD. The results of this study, their contributions to the dysarthria literature, and their

implications for the future direction of research needed will be discussed first. The study

limitations will be discussed next. We will end the discussion by describing ways to

improve the CES, based on input from the study participants.

Hypothesis 1

McHorney described different types of validity that must be addressed in survey

development (McHorney et al., 2002). Some of those types of validity include ecological









validity, social validity, construct validity, and clinical validity, what others call construct

validity (Cook & Campbell, 1979; Maxwell & Delaney 2004; Shadish, Cook, &

Campbell, 2002). We presented the results of two analyses that showed foundational

evidence for the content validity and face validity of the CES for individuals with PD and

a wide range of speech competence in Chapter 2. The purpose of Hypothesis 1 was to

investigate another aspect of validity, construct validity (specifically known-group

construct validity), which is the ability of a measure to discriminate between groups with

and without disease. The result of the test of Hypothesis 1 showed that CES ratings were

significantly higher for healthy, community-dwelling individuals (NO group) than they

were for community-dwelling individuals with dysarthria and Parkinson's disease (PD

group) matched for gender and mean age.

There are many other components of construct validity that remain to be

investigated for the CES. Questions that require further investigation include whether or

not the CES ratings are sensitive to: 1) differences among individuals with PD with

different levels of functional ability; 2) differences among individuals with PD with

different levels of dysarthria severity; 3) differences between individuals with dysarthria

and PD and individuals with PD only, 4) differences between CES ratings when

individuals with PD are in "on" and "off" medication periods, 3) differences between pre-

and post-treatment, and 4) differences that occur as the disease progresses.

Although the NO group (healthy, community-dwelling individuals) rated their CES

significantly higher than the PD group, they did not judge their communicative

effectiveness very effective 100% of the time (M= 28.68, SD = 2.82). Based on this

finding, we propose that there are factors other than speech that contribute to judgments









about communicative effectiveness. Several factors mentioned by participants included:

feeling ill at ease when speaking with a stranger or at a social event, feeling self-

consciousness when speaking, and the hearing acuity of communication partners.

Determining what factors, other than speech, contribute to communicative effectiveness

is an area that requires further investigation.

Hypothesis 2

It has been documented that individuals with PD underestimate their difficulties, or

are not aware of the severity of their deficits (Marsden, Parkes, & Quinn, 1981; Yorkston

et al., 1994; Yorkston et al., 2004). Reasons for these misperceptions have been

hypothesized to occur because of dementia (Bodis-Wollner, 2003) and frontal executive

dysfunction (Ho et al., 2002). However there is an emerging literature that suggests that

individuals with PD have sensorimotor deficits that contribute to a mismatch between

their actual performance that their judgment of performance (Abbruzzese & Berardelli,

2003; Bodis-Wollner, 2003; Ho, Bradshaw, lansek, & Alfredson, 1999). One example of

this deficit pertinent to the LSVT (discussed earlier) is that individuals with PD perceive

the loudness of their voice differently than age-matched individuals without PD

(Abbruzzese & Berardelli, 2003).

There is also literature that showed that caregivers of individuals with chronic

diseases rated those individuals significantly more disabled than the individuals rated

themselves (Andresen, Vahle, & Lollar, 2001; Doyle et al., 2004; Duncan et al., 2002;

Segal & Schall, 1994; Sneeuw, Aaronson, deHaan, & Limburg, 1997). Hypothesis 2 was

designed to investigate the question of the differences between the ratings of

communicative effectiveness of individuals with dysarthria and PD and the CES ratings

that their significant others gave them. The results of the test of Hypothesis 2 showed









that individuals with dysarthria and PD rated their communicative effectiveness

significantly higher than the significant others rated them. Our results add to the

evidence in the respondent literature that individuals with PD rate their performance

higher than do their significant others. However, with regard to the motor speech

literature, our results from Hypothesis 2 differed from those of Ball and colleagues

(2004), who reported that individuals with ALS and their caregivers did not have

significantly different ratings on the CETI-M (a survey very similar to the CES). We

suggest that the difference in our results lends further evidence to the literature that found

individuals with PD do not (or unable to) recognize their communication deficits.

Our results indicated that obtaining information from the individual with the

communication disorder alone might not give researchers and clinicians the most

accurate information about the individual's communicative effectiveness. Our evidence

holds clinical importance for training the communicative dyad, rather than the individual

with dysarthria and PD alone. This is consistent with the findings of others in the field

who work with individuals with PD and motor speech disorders (Hustad, 1999; Hustad,

Beukelman, & Yorkston, 1998; Yorkston et al., 1994; Yorkston et al., 2004).

If the goal of treatment is improved communicative effectiveness for improved

participation in life roles, and there is a significant difference between the communicative

dyad's ratings of communicative effectiveness (i.e., for the individual with dysarthria and

PD) then training the communicative dyad should begin at the earliest stage of treatment

rather than waiting until close to discharge. Such training would provide the clinician,

patient, and significant other an opportunity to try to reach agreement about what has

occurred during treatment and what more needs to be done. However, we need evidence









to demonstrate the benefit of such a treatment paradigm. A future research project could

evaluate the differences in CES ratings between individuals with dysarthria and PD and

significant others who received treatment of the dyad, compared to a group where the

individual with dysarthria and PD received treatment individually and the significant

other received the traditional family education information.

Our research was not designed to determine whether the difference found between

ratings of the PD group and the SO group was due to decreased insight, increased frontal

dysfunction, or the proxy effect. In establishing our inclusion/exclusion criteria we

attempted to control for depression (clients from the MDC are routinely treated for

depression), dementia (MMSE), and apathy (Apathy Scale). However, we did not

address the possibility that other sorts of frontal executive dysfunction could be present.

An area for future investigation would be to determine what role frontal dysfunction (if it

is present) contributes to the CES ratings of individuals with dysarthria and PD. This

kind of research would be an example of what the World Health Organization envisioned

for rehabilitation research when conceptualizing the ICF (WHO, 2001). The WHO ICF

recommends research that spans the entire model, body function and structure

(identifying frontal dysfunction), activity limitations (identifying the deficits that result

from frontal dysfunction), and participation restrictions (identifying the difficulty

completing life roles because of the frontal dysfunction).

Hypothesis 3

The speech intelligibility literature has documented a number of threats to the

internal validity of a study that uses speech intelligibility as a dependent variable. Such

threats include listener familiarization (Hustad & Cahill, 2003; Spitzer et al., 2000),

length of utterance (Spitzer et al., 2000), speaking task (Kempler & Van Lancker, 2002),









medication fluctuations (Goberman & Blomgren, 2003; Goberman et al., 2002; Larson et

al., 1994; Mawdsley & Gamsu, 1971; Quaglieri & Celesia, 1977), practice (Inzana et al.,

1996), and attention (Ho et al., 2002). We attempted to control for these threats when

we developed Hypotheses 3, by stipulating how our speech samples would be obtained,

who would transcribe them, what tasks would be used, when the speech samples would

be collected (i.e., when the individual was in the "on" state of medications). In addition,

we stipulated the level of practice we would let the speakers do before recording the

speech samples. Finally we attempted to control for attentional deficits by conducting the

study in the quietest room of an individual's home. However, the results of Hypothesis 3

were not as we had hypothesized. Neither SIT nor SSI accounted for a significant portion

of the variance in the CES in our initial multiple regression analysis.

Review of the data led us to postulate that the results of the multiple regression

analysis were not significant because the variability within the SIT and SSI scores was

small (individuals in the study had high SIT and SSI scores) although the data met the

statistical assumptions required for regression analysis. Of note, the SIT approached

significance in the initial multiple regression model (SIT and SSI only). It is possible that

with a larger sample SIT would have proven significant for predicting change in CES.

We also found that in our sample, the two speech intelligibility measures (SIT and SSI)

were redundant in the multiple regression models. Further research will be done to

determine which variable was the best predictor of change in CES.

In the post hoc analysis, using a multiple regression model that accounted for the

two speech intelligibility measures (SIT and SSI), a measure of physical disability

(H&Y), and dysarthria severity (DRS), the variable that accounted for the largest portion









of variability of observations in CES was the H&Y (p = .04), a measure of physical

disability, followed by the SSI (which approached significance atp=.10). The results of

this post hoc analysis gave us evidence that the CES was associated with a measure of

participation (the H&Y) more than the measures of impairment (SIT or SIT).

We stated in the introduction that there is not a one-to-one relationship between

speech intelligibility scores and successful functional communication (Weismer &

Laures, 2002; Yorkston, 1996). Our results supported this literature. High SIT and SSI

scores did not predict how an individual rated his communicative effectiveness in

everyday speaking situations. Rather, based on our results, the rating of physical

disability taken from the H&Y was the only significant predictor of change in CES

ratings.

We recognized at the time of data analysis that our PD group was made up of

individuals who had less physical impairment and dysarthria severity based on H&Y

ratings and DRS ratings. We also recognized that the PD group had SIT and SSI scores

that were relatively high (M= .94 and M= .95 respectively). In the future studies should

be designed so that each level of severity is equally represented, and the numbers of

participants in each group are large enough to ensure the power of our results are

adequate to ensure the statistical rigor of our results.

The most important implication of the finding of Hypothesis 3 was that individuals

with dysarthria and PD rated their communicative effectiveness less effective on

everyday speaking situations although their speech intelligibility scores were high. We

propose that this result demonstrated that the CES was tapping a different component of

communication than the one tapped by either of the sentence intelligibility scores. We









suggest that this is evidence that the SIT and the SSI are measures of activity (as they

have been presented to be) and the CES is a measure of communication in the societal

participation component of the ICF. Some might argue that an individuals with a speech

intelligibility score of 95% should not be a candidate for speech therapy. However, if an

individual was able to demonstrate decreased communicative effectiveness on the CES,

that kept him from fulfilling his life roles in societal participation, he might well be a

candidate for speech therapy. We are in the early states of survey development trying to

determine the validity and reliability of the CES; however another area that remains to be

studied in the future is that of the sensitivity and specificity of the CES to identify those

individuals whose communicative effectiveness is limiting their participation in fulfilling

their life roles.

Limitations of the Study

This study was not without limitations. The most obvious limitation, based on our

post hoc analysis was that there are other factors that need to be taken into account

besides an individual's speech when attempting to measure communicative effectiveness,

such as physical ability. Additionally, we received input from both healthy community-

dwelling individuals and community-dwelling individuals with PD and their SOs that

other factors affected their judgments of how they spoke, including being ill-at-ease in

speaking situations, being self-conscious about PD, being uncomfortable around

strangers, being put on the spot, and having conversational partners who were hard of

hearing, and/ or impatient.

Next, our SIT and SSI measures were redundant variables in the multiple

regression models we devised. In the first multiple regression analysis model SIT









approached significance (p = .10) and in the second (when severity was added into the

model) the SSI approached significance (p = .10).

In addition to the problem of redundancy, the literature presents conflicting reports

on what role a spontaneous speech sample plays in determining speech intelligibility

versus the traditional intelligibility measure where words and sentences are read by the

speaker. Some authors report improved intelligibility because the listener an derive

meaning from the context of what is being said (Dagenais, Garcia, & Watts, 1998;

Dagenais, Watts, Turnage, & Kennedy, 1999; Hustad & Beukelman, 2002; Hustad &

Beukelman, 2001; Hustad & Cahill, 2003). Others have found that spontaneous speech

samples result in decreased intelligibility ratings (Kempler & Van Lancker, 2002). In our

sample, the SIT and SSI tasks were redundant, which could lead one to suggest choosing

just one speech intelligibility task. However, we also reviewed the transcription

methodology we chose to determine if it would point to methodological problems we had

overlooked.

A review of the inter-rater reliability on the SSI, although still good, was 10%-15%

less than that of the SIT. Refer to Table 3-6. We used the AIDS transcription

instructions for both intelligibility tasks (Yorkston et al., 1984). A key of correct

sentences was available to check against the transcribed SIT samples. However, there

were no keys with which to compare the SSI transcriptions. In the future, if a

spontaneous speech sample is used, the PI could transcribe the SSI samples immediately

after they are collected while the content of the sample was fresh. That transcription

could serve as the key with which to compare the transcribed samples. This method









might reduce the amount of discrepancy between the intelligible words and unintelligible

words in the SSI samples.

Improving the Communicative Effectiveness Survey

Survey development typically involves getting input from individuals with the

problem being studied and their significant others through focus groups or individual

interviews. Such input gives ecological/social validity to a survey (Antonius et al., 1995;

Lomas et al., 1989; Yorkston et al., 1994; Yorkston et al., 2001), which in turn should

increase the likelihood of content validity and face validity of an instrument. Lomas and

colleagues' (1989) designed the Communicative Effectiveness Index (CETI) to measure

changes in the functional communication of individuals with aphasia. They used input

from individuals with aphasia and caregivers during CETI development. Yorkston and

colleagues (1999) modified the CETI to measure changes in the functional

communication of individuals with motor speech disorders (such as the dysarthria

associated with PD). As with the Lomas group, Yorkston and colleagues (1999) gathered

input from individuals with a variety of motor speech disorders before modifying the

CETI and called it the CES (the measure used in our study).

Because individuals with motor speech disorders had already given input into the

survey, we did not use focus groups ahead of time but rather asked for feedback from the

individuals with dysarthria and PD and their significant others after they had completed

the CES. There were only a few addition items suggested. In the future, we will consider

adding items to the instrument based on the input from the participants. Once items are

added, further analysis would be made to ensure the stability of the measurement

properties of the CES.









We believe the information participants gave in response to the third debriefing

question "is there any other information about your communicative effectiveness that you

think other healthcare professionals should know?" clearly showed that individuals with

PD and their SO's viewed communicative effectiveness as a societal participation

measure. Examples of their responses included: feeling very self-conscious about one's

speech, having trouble communicating when being put on the spot, feeling tense made

speech worse, being perceived as cognitively impaired because it is difficult to respond

quickly. See Appendix C for a full listing of comments to all questions.

The groups' responses also encompassed more than communication. Other things

they thought healthcare professionals should know included: fear about what the future

holds, disease progression, ability to remain independent, the strain the disease puts on

marriage and family, personal safety, financial security, physical and emotional ups and

downs, role reversals in relationships, stigma of PD (tremors, reduced mobility, using a

wheelchair, slowness of thinking and responding), and impatience of the public.

Many of the concerns expressed had little to do with speech per se, but had much to

do with maintenance of relationships and independence, both of which require certain

levels of communicative effectiveness. Uncovering the impact that decreased

communicative effectiveness has on those living with PD (both patient and significant

others) is an area that is ripe for further research. This kind of research (qualitative

research) is emerging in the aphasiology literature (Damico & Simmons-Mackie, 2003;

Kagan & LeBlanc, 2002; Lyon, 2000), but there is very little of it being systematically

done in the area of motor speech disorders in general, and hypokinetic dysarthria of

Parkinson's disease in particular (Antonius et al., 1995; Yorkston et al., 1994).









Future research questions that are of particular interest relative to doing qualitative

research in the area of communicative effectiveness of individuals with dysarthria and PD

include: 1) what does the phrase communicative effectiveness mean to you? 2) what are

the biggest societal barriers you face to communicate effectively? 3) how do medication

fluctuations affect your communicative effectiveness? 3) do you manage your

medications to maximize communicative effectiveness in societal participation and if so,

how? 4) do you manage your medications for other kinds of societal participation, and if

so, how? Answers to these questions would be beneficial in further establishing the

validity of communicative effectiveness as a measure of societal participation.

Conclusion

In conclusion, this study has provided evidence that adds to the face validity,

content validity, and construct validity of the CES as a measure of communication for the

societal participation component the WHO ICF model. The CES differentiated healthy,

community-dwelling elders from gender- and mean-age-matched individuals with

dysarthria and PD. In addition, we showed that there was a significant difference

between the way individuals with dysarthria and PD rated their communicative

effectiveness and the way their significant others rated the communicative effectiveness

of the individual with PD. Finally, we found evidence that individuals with dysarthria

and PD rated their communication less effective, as their physical ability declined

although their sentence intelligibility scores and spontaneous speech intelligibility scores

remained high.

In motor speech disorders, we have many measures of body functions/structures

(perceptual judgments, acoustic, and physiologic measures) and activity (speech

intelligibility), but we have yet to develop a measure of communication in societal









participation. Such a measure is needed by speech therapists to demonstrate that the

treatments they administer at the clinical level result in success for the client where the

client needs it, to fulfill life roles at home, at work in the community. Such measures are

also needed by researchers interested in motor speech disorders, so that as we go about

designing studies to demonstrate the efficacy and effectiveness of new treatments we

remember the goal of speech therapy is to enable a person to communicate more

effectively to participate in his life roles.

Although there are many questions left to be answered regarding the reliability and

other types of validity of the CES for individuals with dysarthria and PD, we believe we

have presented evidence that the CES holds promise as a measure of communication for

societal participation to fulfill life roles.















APPENDIX A
COMMUNICATIVE EFFECTIVENESS SURVEY












NAME:


In this survey we ask you to rate how effective your speech is in different
communication situations. Please read each statement. Then rate how effectively you
communicate in that situation. If you feel your speech is very effective, mark the 4. If
your speech does not allow you to communicate at all in a situation, mark the 1. Feel free
to use any number on the scale.

1. Having a conversation with a family member or friends at home.
Not able to do Very Effective
1 2 3 4

2. Participating in conversation with strangers in a quiet place.
Not able to do Very Effective
1 2 3 4


3. Conversing with a familiar person over the telephone.
Not able to do


Very Effective


1 2 3 4

4. Conversing with a stranger over the telephone.
Not able to do Very Effective
1 2 3 4

5. Being part of a conversation in a noisy environment (social gathering).
Not able to do Very Effective
1 2 3 4

6. Speaking to a friend when you are emotionally upset or you are angry.
Not able to do Very Effective
1 2 3 4

7. Having a conversation while traveling in a car.
Not able to do Very Effective
1 2 3 4

8. Having a conversation with someone at a distance (across a room).
Not able to do Very Effective
1 2 3 4

Hustad, K.C., 1999. Optimizing Communicative Effectiveness: Bringing it together. In Yorkston, K.M.,
Beukelman, D.R., Strand, E.A, & Bell, K.R. (Eds.), Management of Motor Speech Disorders (2nd ed.),
(pp.438-541).


DATE:















APPENDIX B
HOEHN AND YAHR STAGING FOR PARKINSON'S DISEASE

Name: Date:

1. Stage One

1. Signs and symptoms on one side only
2. Symptoms mild
3. Symptoms inconvenient but not disabling
4. Usually present with tremor of one limb
5. Friends have noticed changes in posture, locomotion and facial expression

2. Stage Two
1. Symptoms are bilateral
2. Minimal disability
3. Posture and gait affected

3. Stage Three
1. Significant slowing of body movements
2. Early impairment of equilibrium on walking or standing
3. Generalized dysfunction that is moderately severe

4. Stage Four
1. Severe symptoms
2. Can still walk to a limited extent
3. Rigidity and bradykinesia
4. No longer able to live alone
5. Tremor may be less than earlier stages

5. Stage Five
1. Cachectic stage
2. Invalidism complete
3. Cannot stand or walk
4. Requires constant nursing care


Hoehn, M.M. & Yahr, M.D. (1967). Parkinsonism: onset, progression, and mortality.
Neurology, 17(5): 427-442.















APPENDIX C
PARTICIPANT FEEDBACK ON THE CES









Table C-1. Participant feedback
Question Individuals with Dysarthria and PD
1. Overall, what is your -Succinct, clear
impression of the survey? -Good and short
-Short is a good thing
-I like short tests
-Short & sweet
-You need more numbers. I would
have liked to put some in-between
-Simple and understandable
-Is this enough questions to be
meaningful?
-I am neutral on it
-It was quick and easy
-I'm benign on it
-Good and short
-I may be a 3.5 on some


2. Are there other social
speaking situations we
should add to this survey?


-Perhaps a question about
conversing for a lengthy period of
time
-I don't see anything about my
speech when I am tired
-Outdoors is a problem, I can't
yell like I used to
-Hard to order food at a restaurant
with the noise and the rush
-Maybe something about talking
faster-that is my wife's complaint
-Most of my friends are hard of
hearing. I think that is the
problem.
-I don't answer the phone
anymore so that question doesn't
really apply to me.
-Add something about fatigue
level and speaking.
-You need a question about
talking when you are in an
anxious situation, or being judged,
not just mad or whatever.
-Change the emotionally upset
question. I don't let myself get
mad. Maybe use the term
emotionally involved.
-It is hard to talk and dance and
we love to dance.


Significant Others
-Some of the wording is not
clear
-It doesn't really apply to
most of our problems with
lifestyle changes
-Very lifelike and applicable











-I worry when he is outdoors
by himself he can't yell at
me for help.
-Maybe ask about talking at
home with the TV on or
when there are other people
here-it makes a difference.
-When all the grandchildren
are here I can't hear him.
Maybe one about talking
when he is tired.
-When more than two of us
are here at home it is hard to
hear him and he cannot seem
to get into the
conversation-it is as though
we pass him by.










Table C-l continued.
Question
3. Is there any other
information about your
communicative
effectiveness that you
think other healthcare
professionals should
know?


Individuals with Dysarthria and PD
-I notice my voice is quieter, but I
don't think it makes much
difference, does it?
-I am newly diagnosed but worry
about the future and what is going
to happen.
-Nobody realizes how much I
fluctuate from day to day, not
because of meds, but just how I
feel in general
-I have trouble speaking when I
am put on the spot for an answer
-People seem to think I am more
cognitively impaired because of
my slowness to respond
- Everyone notices how clear my
speech is when I am mad
-Everything is harder now
-Exercise helps my fatigue
-My voice is worse when I get
confused, like when I am tired.
-I want everyone to know that
therapy helps. I think people
interact differently with me
because I come across as a
competent communicator now and
I speak up.
-Feeling tense makes my speech
worse
-The hearing of the listener is very
important
-I have become very self-
conscious about my speech
-My husband won't go and leave
me alone although I want him to
because I can take care of myself
very well.
-I don't want to be a burden to my
wife, but I already am
-It is hard on the person I am with.
I have lost my facial expression
and he can't tell when I am happy.
-Everything takes a long time to
do and people aren't always
patient.


Significant Others
-No one really addresses
caregiver burden. I am
trapped here with him.
-He knows everyone around
here and is embarrassed by
some of his problems. No
one has talked to us about
that kind of thing.
-PD is very hard on a
marriage and nobody tells
you that.
-The fear of what is coming
is almost unbearable.
-I feel desperate.
-I take Prozac to manage my
feelings of dread for what is
ahead.
-We are both getting older
and our kids are up north.
We don't have much help.
-I have to do all kinds of
things I never thought I
would do.
-I don't like making
decisions by myself, but
sometimes I just do.
-It takes us so long to do
anything that it is easier to
stay home.
















LIST OF REFERENCES


Abbruzzese, G., & Berardelli, A. (2003). Sensorimotor integration in movement
disorders. Movement Disorders, 18(3), 231-240.

Adams, S. G. (1997). Hypokinetic dysarthria in Parkinson's disease. In M. R. McNeil
(Ed.), Clinical Management of Sensorimotor Speech Disorders (pp. 2261-2285).
New York: Thieme.

Andresen, E., Vahle, V., & Lollar, D. (2001). Proxy reliability: health-related quality of
life measures for people with disability. Quality ofLife Research, 10(7), 609-619.

Ansel, B. M., & Kent, R. D. (1992). Acoustic-phonetic contrasts and intelligibility in the
dysarthria associated with mixed cerebral palsy. Journal of Speech andHearing
Research, 35, 296-308.

Antonius, K., Beukelman, D. R., & Reid, R. (1995). Communication disability of
Parkinson's disease. In D. A. Robin, K. M. Yorkston & D. R. Beukelman (Eds.),
Disorders of Motor Speech: recent advances in assessment, treatment, and
clinical characterization (pp. 2275-2286). Baltimore: Paul H. Brooks.

Ball, L. J., Beukelman, D. R., & Pattee, G. L. (2004). Communication effectiveness of
individuals with amyotrophic lateral sclerosis. Journal of Communication
Disorders, 37, 197-215.

Bergner, M., Bobbitt, R. A., Carter, W. B., & Gilson, B. S. (1981). The sickness impact
profile: development and final revision of a health status measure. Medical Care,
19, 787-805.

Beukelman, D. R., & Yorkston, K. M. (1977). A communication system for the severely
dysarthric speaker with an intact language system. Journal of Speech andHearing
Disorders, 42, 265-270.

Beukelman, D. R., & Yorkston, K. M. (1979). The relationship between information
transfer and speech intelligibility of dysarthric speakers. Journal of
Communication Disorders, 12, 189-196.

Bodis-Wollner, I. (2003). Neuropsychologican and perceptual defects in Parkinson's
disease. Parkinsonism and Related Disorders, 9, S83-S89.

Bond, T. G., & Fox, C. M. (2001). Applying the Rasch Model: Fundamental
measurement in the human sciences. Mahwah, NJ: Lawrence Erlbaum Associates.









Brandt, E. N., Jr., & Pope, A. M. (1997). Models of disability and rehabilitation. In E. N.
Brandt, Jr. & A. M. Pope (Eds.), Enabling America: assessing the role of
rehabilitation science and engineering (pp. 24-39). Washington, D.C.: National
Academy Press.

Cardol, M., deJong, B. A., van den Bos, G., Beelen, A., deGroot, I. J. M., & deHaan, R.
J. (2002). Beyond disability: perceived participation in people with a chronic
disabling condition. Clinical Rehabilitation, 16, 27-35.

Cook, T.D. & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis for
field settings. Chicago: Rand McNally.

Dagenais, P. A., Garcia, J. M., & Watts, C. R. (1998). Acceptability and intelligibility of
mildly impaired dysarthric speech by different listeners. In M. P. Cannito, K.
Yorkston & D. Beukelman (Eds.), Neuromotor Speech Disorders: nature,
assessment, and management (pp. 213-227). Baltimore: Paul H. Brookes.

Dagenais, P. A., Watts, C. R., Turnage, L. M., & Kennedy, S. (1999). Intelligibility and
acceptability of moderately dysarthric speech by three types of listeners. Journal
of Medical Speech-Language Pathology, 7, 91-96.

Damico, J. S., & Simmons-Mackie, N. N. (2003). Qualitative research and speech-
language pathology: A tutorial for the clinical realm. American Journal of
Speech-Language Pathology, 12, 131-143.

Darley, F. L., Aronson, A. E., & Brown, J. R. (1975). Motor Speech Disorders.
Philadelphia: W.B. Saunders.

Doyle, P., McNeil, M. R., JM, M., Prieto, L., Hula, W., Lustig, A., Ross, K., Wambaugh,
J., Rothi, L., & Elman, R. (2004). The Burden of Stroke Scale (BOSS) provides
valid and reliable score estimates of functioning and well-being in stroke
survivors with and without communication disorders. Journal of Clinical
Epidemiology, 57(10), 997-1007.

Duffy, J. R. (1995). Motor Speech Disorders: substrates, differential diagnosis, and
management. St. Louis: Mosby.

Duffy, J. R. (2005). Motor Speech Disorders: Substrates, differential diagnosis, and
management (2nd ed.). St. Louis, MO: Elsevier.

Duncan, P., Lai, S., Tyler, D., Perera, S., Reker, D., & Studenski, S. (2002). Evaluation
of proxy responses to the stroke impact scale. Stroke, 33(11), 2593-2599.

Enderby, P. (2000). Assessment and treatment of functional communication in dysarthria.
In L. E. Worrall & C. M. Frattali (Eds.), Neurogenic Communication Disorders: a
functional approach (pp. 247-261). New York: Thieme.









Faul, F., & Erdfelder, E. (1996). G*Power: a general power analysis program. Behavior
Research Methods, Instruments, & Computers, 28, 1-11.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state. A practical
method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12, 189-198.

Gilson, B. S., Gilson, J. S., Bergner, M., Bobbitt, R. A., Kressel, S., Pollard, W. E., &
Vesselago, M. (1975). The sickness impact profile development of an outcome
measure of health care. American Journal of Public Health, 65, 1304-1310.

Goberman, A. M., & Blomgren, M. (2003). Parkinsonian speech disfluencies: effects of
1-dopa-related fluctuations. Journal of Fluency Disorders, 28(1), 55-70.

Goberman, A. M., Coelho, C., & Robb, M. (2002). Phonatory characteristics of
Parkinsonian speech before and after morning medication: the ON and OFF
states. Journal of Communication Disorders, 35, 217-239.

Hammen, V. L., & Yorkston, K. (1996). Speech and pause characteristics following
speech rate reduction in hypokinetic dysarthria. Journal of Communication
Disorders, 29, 429-445.

Ho, A. K., Bradshaw, J. L., lansek, R., & Alfredson, R. (1999). Speech volume regulation
in Parkinson's disease: effects of implicit cues and explicit instructions.
Neuropsychologia, 37, 1453-1460.

Ho, A. K., lansek, R., & Bradshaw, J. L. (2002). The effect of a concurrent task on
parkinsonian speech. Journal of Clinical and Experimental Neuropsychology,
24(1), 36-47.

Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: onset, progression, and mortality.
Neurology, 17(5), 427-442.

Hustad, K. C. (1999). Optimizing communicative effectiveness: bringing it together. In
K. Yorkston, D. Beukelman, E. A. Strand & K. R. Bell (Eds.), Management of
Motor Speech Disorders in Children andAdults (pp. 483-537). Austin, TX: Pro-
ed.

Hustad, K. C., & Beukelman, D. (2002). Listener comprehension of severely dysarthric
speech: effects of linguistic cues and stimulus cohesion. Journal of Speech and
Hearing Research, 45, 545-558.

Hustad, K. C., & Beukelman, D. R. (2001). Effects of linguistic cues and stimulus
cohesion on intelligibility of severely dysarthric speech. Journal of Speech and
Hearing Research, 44, 497-512.

Hustad, K. C., Beukelman, D. R., & Yorkston, K. M. (1998). Functional outcome
assessment in dysarthria. Seminars in Speech and Language, 19(3), 291-302.









Hustad, K. C., & Cahill, M. A. (2003). Effects of presentation mode and repeated
familiarization on intelligibility of dysarthric speech. American Journal of
Speech-Language Pathology, 12, 198-206.

Inzana, C., Driskell, J., Salas, E., & Johnson, J. (1996). Effects of preparatory
information on enhancing performance under stress. Journal ofApplied
Psychology, 81, 429-435.

Jette, A. M. (2003). Assessing disability in studies on physical activity. American Journal
ofPreventative Medicine, 25(3), 122-128.

Jette, A. M., Haley, S. M., & Kooyoomjian, J. T. (2003). Are the ICF activity and
participation dimensions distinct? Journal ofRehabilitation Medicine, 35(3), 145-
149.

Kagan, A., & LeBlanc, K. (2002). Motivating for infrastructure change: toward a
communicatively accessible, participation-based stroke care system for all those
affected by aphasia. Journal of Communication Disorders, 35, 153-169.

Kempler, D., & Van Lancker, D. (2002). Effect of speech task on intelligibility in
dysarthria: a case study of Parkinson's disease. Brain and Language, 80, 449-464.

Kent, R. D. (2000). Research on speech motor control and its disorders: a review and
prospective. Journal of Communication Disorders, 33(5), 391-428.

Kent, R. D., & Read, C. (1992). The Acoustic Analysis of Speech. San Diego: Singular
Publishing Group.

Kent, R. D., Vorperian, H. K., Kent, J. F., & Duffy, J. R. (2003). Voice dysfunction in
dysarthria: application of the Multi-Dimensional Voice Program(tm). Journal of
Communication Disorders, 36, 281-306.

Keysor, J. J., & Jette, A. M. (2001). Have we oversold the benefit of late-life exercise?
Journal of Gerontology: Medical Sciences, 56A(7), M412-M423.

Klasner, E. R., Yorkston, K. M., & Strand, E. A. (1999). Patters of perceptual features in
speakers with ALS: a preliminary study of prominence and intelligibility
considerations. Journal of Medical Speech-Language Pathology, 7(2), 117-125.

Larson, K., Ramig, L., & Scherer, R. (1994). Acoustic and glottographic voice analysis
during drug-related fluctuations in Parkinson disease. Journal of Medical Speech-
Language Pathology, 2, 227-239.

Linacre, J. M. (1991). Winsteps Ministep Rasch-model computer programs [computer
software].

Linacre, J. M. (1994). Sample size and item calibration stability. Rasch Measurement
Transactions, 7(4), 328.









Linacre, J. M. (2002). Optimizing rating scale category effectiveness. Journal of Applied
Measurement, 3(1), 85-106.

Lomas, J., Pickard, L., Bester, S., Elbard, H., Finlayson, A., & Zoghaib, C. (1989). The
communicative effectiveness index: development and psychometric evaluation of
a functional communication measure for adult aphasia. Journal of Speech and
Hearing Disorders, 54(1), 113-124.

Lyon, J. G. (2000). Finding, defining, and refining functionality in real life for people
confronting aphasia. In L. E. Worrall & C. M. Frattali (Eds.), Neurogenic
Communication Disorders: a functional approach (pp. 137-161). NY: Thieme.

Marin, R. S. (1990). Differential diagnosis and classification of apathy. American Journal
ofPsychiatry, 147, 22-30.

Marsden, C. D., Parkes, J., & Quinn, N. (1981). Fluctuations of disability in Parkinson's
disease-clinical aspects. In C. D. Marsden & S. Fahn (Eds.), Movement Disorders
(pp. 96-122). London: Butterworth.

Mawdsley, C., & Gamsu, C. (1971). Periodicity of speech in Parkinsonism. Nature, 231,
315-316.

Maxwell, S.E. & Delaney, H.D. (2004). Designing Experiments and Analyzing Data (2nd
ed). Mahwah, NJ: Lawrence Erlbaum Associates.

McHorney, C. A., Robbins, J., Lomax, K., Rosenbek, J. C., Chignell, K. A., & Kramer,
A. E. (2002). The SWAL-QOL and SWAL-CARE outcomes tool for
oropharyngeal dysphagia in adults: III. Documentation of reliability and validity.
Dysphagia, 17, 97-114.

Nordenfelt, L. (2003). Action theory, disability and ICF. Disability and Rehabilitation,
25(18), 1075-1079.

Parkinson's Disease Society website (2005). URL: http://www.parkinsons.org.uk. Last
accessed 04/16/05.

Perenboom, R. J. M., & Chorus, A. M. J. (2003). Measuring participation according to
the ICF. Disability and Rehabilitation, 25(11-12), 577-587.

Pollard, B., & Johnston, M. (2001). Problems with the sickness impact profile: a
theoretically based analysis and a proposal for a new method of implementation
and scoring. Social Science and Medicine, 52, 921-934.

Quaglieri, C., & Celesia, G. (1977). Effect ofthalamotomy and levodopa therapy on the
speech of Parkinson patients. European Journal ofNeurology, 15, 34-39.









Ramig, L. 0. (1994). Voice treatment for patients with Parkinson's disease: development
of an approach and preliminary efficacy data. Journal of Medical Speech-
Language Pathology, 2(3), 191-209.

Ramig, L. 0. (1995). Comparison of two forms of intensive speech treatment for
parkinson disease.

Ramig, L. O., Countryman, S., O'Brien, C., Hoehn, M., & Thompson, L. (1996).
Intensive speech treatment for patients with Parkinson's disease: short- and long-
term comparison of two techniques. Neurology, 47(6), 1496-1504.

Ramig, L. O., Countryman, S., Thompson, L., & Horii, Y. (1993). Comparison of two
forms of intensive speech treatment for Parkinson disease. Journal of Speech and
Hearing Research, 38(5), 1232-1251.

Ramig, L. O., Countryman, S., Thompson, L. L., & Horii, Y. (1995). Comparison of two
forms of intensive speech treatment for Parkinson disease. Journal of Speech and
Hearing Research, 38(6), 1232-1235.

Ramig, L. O., Sapir, S., Countryman, S., Pawlas, A., O'Brien, C., Hoehn, M., &
Thompson, L. L. (2001). Intensive vice treatment (LSVT) for patients with
Parkinson's disease: a 2 year follow up. Journal ofNeurology, Neurosurgery and
Psychiatry, 71, 493-498.

Rasch, G. (1980). Probabilistic models for some intelligence and attainment tests
(Expanded). Chicago: University of Chicago Press.

Rosenbek, J. C., & LaPoint, L. L. (1978). The dysarthrias: description, diagnosis and
treatment. In F. L. Darley (Ed.), Clinical Management ofNeurogenic
Communication Disorders. Boston: Little, Brown and Company.

Segal, M., & Schall, R. (1994). Determining functional/health status and its relation to
disabillity in stroke survivors. Stroke, 25, 2391-2397.

Shadish, W.R., Cook, T.D., & Campbell, D.T. (2002). Experimental and Quasi-
Experimental Designs for Generalized Causal Inference. Boston: Houghton-
Mifflin.

Sneeuw, K., Aaronson, N., deHaan, R. J., & Limburg, M. (1997). Assessing quality of
life after stroke: the value and limitations of proxy ratings. Stroke, 28, 1541-1549.

Spitzer, S. M., Liss, J. M., Caviness, J. N., & Adler, C. (2000). An exploration of
familiarization effects in the perception of hypokinetic and ataxic dysarthric
speech. Journal of Medical Speech-Language Pathology, 8, 285-293.









Starkstein, S. E., Mayberg, H. S., Preziosi, T. J., Andrezejewski, P., Leiguarda, R., &
Robinson, R. G. (1992). Reliability, validity, and clinical correlates of apathy in
Parkinson's disease. Journal ofNeuropsychiatry and Clinical Neurosciences, 4,
134-139.

Urbaniak, G. C., & Pious, S. (1997). Research Randomizer, 2004, from
http://222.randomizer.org

Van Den Eeden, S. K., Tanner, C. M., Bernstein, A. L., Fross, R. D., Leimpeter, A.,
Bloch, D. A., & Nelson, L. M. (2003). Incidence of Parkinson's disease: variation
by age, gender, and race/ethnicity. American Journal ofEpidemiology, 157(11),
1015.

Velozo, C. A., & Peterson, E. W. (2001). Developing meaningful fear of falling measures
for community dwelling elderly. American Journal ofPhysical Medicine and
Rehabilitation, 80(9), 662-673.

Weismer, G., & Laures, J. S. (2002). Direct magnitude estimates of speech intelligibility
in dysarthria: effects of a chosen standard. Journal of Speech andHearing
Research, 45(3), 421-433.

WHO. (2001). International Classification ofFunctioning, Disability and Health: ICF.
Geneva: WHO.

Wright, B. D. (1997). A history of social science measurement. Educational
Measurement: Issues and Practice, 16(4), 33-45.

Wright, B. D., & Linacre, J. M. (1989). Observations are always ordinal; measurements,
however, must be interval. Archives ofPhysical Medicine and Rehabilitation,
70(12), 857-860.

Wright, B. D., & Stone, M. H. (1979). Best Test Design: Rasch Measurement. Chicago,
IL: MESA Press.

Yorkston, K. M. (1996). Treatment efficacy: dysarthria. Journal of Speech andHearing
Research, 39, S46-S57.

Yorkston, K. M., & Beukelman, D. R. (1981 a). Assessment of intelligibility of dysarthric
speech. In. Tigard: CC Publications.

Yorkston, K. M., & Beukelman, D. R. (1981b). Communication efficiency of dysarthric
speakers as measured by sentence intelligibility and speaking rate. Journal of
Speech and Hearing Disorders, 46(3), 296-301.

Yorkston, K. M., Beukelman, D. R., Strand, E. A., & Bell, K. R. (1999). Management of
Motor Speech Disorders in children andAdults (2nd ed.). Austin, TX: Pro-ed.









Yorkston, K. M., Beukelman, D. R., & Tice, R. (1996). Speech Intelligibility Test
[speech analysis] (Version 1.0). Lincoln, NE: Tice Technology Services.

Yorkston, K. M., Beukelman, D. R., & Traynor, C. D. (1984). Assessment of
Intelligibility of Dy )i t//l Ai Speech. Austin, TX: Pro-ed.

Yorkston, K. M., Bombardier, C., & Hammen, V. L. (1994). Dysarthria from the
viewpoint of individuals with dysarthria. In J. Till, K. Yorkston & D. Beukelman
(Eds.), Motor Speech Disorders: advances in assessment and treatment.

Yorkston, K. M., Dowden, P. A., & Beukelman, D. R. (1992). Intelligibility measurement
as a tool in the clinical management of dysarthric speakers. In R. D. Kent (Ed.),
Intelligibility in Speech Disorders: theory, measurement and management (pp.
2265-2286). Amsterdam: John Benjamins.

Yorkston, K. M., Klasner, E. R., & Swanson, K. M. (2001). Communication in context: a
qualitative study of the experiences of individuals with multiple sclerosis.
American Journal of Speech-Language Pathology, 5, 126-137.

Yorkston, K. M., Miller, R. M., & Strand, E. A. (2004). Management of Speech and
Swallowing in Degenerative Diseases (2nd ed.). Austin, TX: Pro-ed.

Yorkston, K. M., Strand, E. A., & Kennedy, M. R. T. (1996). Comprehensibility of
dysarthric speech: implications for assessment and treatment planning. American
Journal of Speech-Language Pathology, 5, 55-66.















BIOGRAPHICAL SKETCH

Neila J. Donovan received her Bachelor of General Studies degree with a major in

speech pathology and minors in education and audiology from the University of Kansas,

Lawrence, Kansas in January 1983. She received her Master of Arts degree in speech-

language pathology in July 1985, also from the University of Kansas. She currently

holds the Certificate of Clinical Competency from the American Speech-Language-

Hearing Association. She plans to remains active in both clinical practice and clinical

research to improve the communicative effectiveness of individuals with acquired motor

speech disorders after graduation.